U.S. patent application number 14/025142 was filed with the patent office on 2014-01-09 for abrasive compact.
This patent application is currently assigned to Element Six Abrasives S.A.. The applicant listed for this patent is Element Six Abrasives S.A.. Invention is credited to Mathias HERRMANN, Kekeletso MLUNGWANE, Iakovos SIGALAS.
Application Number | 20140007515 14/025142 |
Document ID | / |
Family ID | 39952369 |
Filed Date | 2014-01-09 |
United States Patent
Application |
20140007515 |
Kind Code |
A1 |
SIGALAS; Iakovos ; et
al. |
January 9, 2014 |
ABRASIVE COMPACT
Abstract
The invention relates to an abrasive compact comprising a mass
of diamond particles and a silicon containing binder phase wherein
the diamond particles are present in an amount less than 75 volume
% and the binder phase contains less than 2 volume % unreacted
(elemental) silicon. The invention further relates to a method of
producing an abrasive compact including the steps of forming a feed
diamond powder into a diamond preform, interposing a separating
mechanism between the diamond preform and a silicon infiltrant
source, heating the diamond preform and silicon infiltrant source
until the infiltrant is molten and the preform and infiltrant are
isothermal and allowing infiltration from the molten silicon
infiltrant source to occur into the diamond preform.
Inventors: |
SIGALAS; Iakovos; (Springs,
ZA) ; MLUNGWANE; Kekeletso; (Springs, ZA) ;
HERRMANN; Mathias; (Springs, ZA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Element Six Abrasives S.A. |
Luxembourg |
|
LU |
|
|
Assignee: |
Element Six Abrasives S.A.
Luxembourg
LU
|
Family ID: |
39952369 |
Appl. No.: |
14/025142 |
Filed: |
September 12, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12670229 |
Aug 25, 2010 |
8562702 |
|
|
PCT/IB2008/052953 |
Jul 23, 2008 |
|
|
|
14025142 |
|
|
|
|
60951290 |
Jul 23, 2007 |
|
|
|
Current U.S.
Class: |
51/307 |
Current CPC
Class: |
C04B 2235/5481 20130101;
C04B 2235/80 20130101; C04B 2235/616 20130101; C04B 2235/48
20130101; C04B 2235/96 20130101; C04B 2235/5436 20130101; C04B
35/62839 20130101; C04B 35/52 20130101; C04B 2235/9623 20130101;
C04B 35/645 20130101; C04B 2235/428 20130101; C04B 2235/6562
20130101; C04B 2235/608 20130101; B24D 3/14 20130101; C04B 2235/728
20130101; C04B 35/6269 20130101; C04B 35/573 20130101; C04B
2235/3826 20130101; C04B 2235/427 20130101; C04B 2235/721 20130101;
C04B 35/6316 20130101; F41H 5/0492 20130101 |
Class at
Publication: |
51/307 |
International
Class: |
B24D 3/14 20060101
B24D003/14 |
Foreign Application Data
Date |
Code |
Application Number |
Jul 23, 2007 |
ZA |
2007/06085 |
Claims
1. An abrasive compact comprising a mass of diamond particles and
binder phase comprising silicon, wherein the diamond particles have
a particle size of at least 5 microns and are present in an amount
less than 70 volume % and the binder phase ucomprises no detectable
free or elemental silicon determined by XRD.
2. The compact according to claim 1, wherein the diamond particles
are present in an amount of more than 5 to volume %.
3. (canceled)
4. (canceled)
5. The compact according to claim 1, wherein the diamond particles
have an average grain size less than 10 .mu.m.
6. The compact according to claim 1, wherein the diamond particles
are present in an amount of more than 20 volume %.
7. The compact according to claim 1, wherein the diamond particles
are present in an amount of more than 30 volume %.
8. The compact according to claim 1, wherein the diamond particles
are present in an amount of more than 40 volume %.
9. The compact according to claim 1, wherein at least some of the
silicon in the binder phase is present as silicon carbide.
10. The compact according to claim 9, wherein a majority of the
silicon is present in the binder phase as silicon carbide.
11. The compact according to claim 9, wherein the silicon carbide
is microcrystalline in nature.
12. The compact according to claim 10, wherein the silicon carbide
is microcrystalline in nature.
Description
CROSS REFERENCE TO RELATED APPLICATION(S)
[0001] This application is a continuation of U.S. application Ser.
No. 12/670,229, filed on Aug. 25, 2010, the entire content of which
is incorporated herein by reference.
INTRODUCTION
[0002] This invention relates to a diamond compact body comprising
diamond particles bonded together by a silicon-containing binder or
bonding phase. Such compacts are well-known in the art and are
useful as an abrasive, cutting tool, nozzle or other wear-resistant
part. The invention extends to a method of manufacturing such a
diamond compact body.
BACKGROUND OF THE INVENTION
[0003] Diamond is the hardest material known to man. Because of
this, it finds extensive industrial application where ultra-hard
material properties are needed. Due to its high hardness, it is
difficult to make diamond tools of different shapes and sizes
purely from cutting and shaping diamond. This has led to the
development of diamond composite materials which consist of small
diamond grains either sintered together through a liquid phase
sintering process, or held together in a matrix by a binder phase
material. The former process gives rise to the class of
polycrystalline diamond materials (PCD), while the latter results
in a number of composite materials, of which the foremost is that
of SiC-diamond composites. The introduction of the second phase
improves the formability and the fracture toughness of such
diamond-based material.
[0004] Metallic phases such as cobalt are present in PCD and are
commonly used as liquid phase sintering aids in the production of
that material. These metals however were found to catalyse the
graphitization of diamond thus limiting the application
temperatures of these PCD materials to below 1000.degree. C.
Silicon carbide has been found to be exceptionally good as a
diamond binder phase. Because of the structural similarities
between diamond and silicon carbide, a strong bond forms between
them that results in a material with very strong adhesion between
the diamond grains and the SiC matrix. SiC is commonly formed in
situ from the reaction between diamond and/or amorphous carbon or
graphite with silicon. SiC does not react with diamond and hence
the composite material can be used at temperatures above
1000.degree. C. However, the application temperature may be limited
by the melting temperature of silicon if some unreacted silicon is
present in the final product.
[0005] There are two different generic routes of production of
these composites: [0006] mixing of a powdered silicon source with
diamond particles and densification of the mixture under pressure
with temperature (reaction sintering), or [0007] infiltration by a
silicon-containing melt of a preform made from diamond powder or
from mixtures of diamond with graphite or resin.
[0008] Reaction sintering to obtain fully dense compacts is only
relatively straightforward under the high-pressure high-temperature
(HpHT) conditions typically associated with diamond synthesis.
Under low pressure conditions (such as Hot Pressing (HP) and Hot
Isostatic Pressing (HIP)), the volume decrease associated with the
local formation of SiC from the intermingled silicon source and
diamond may well result in residual porosity. Therefore a pressure
high enough for densification of the reacted compact such as
diamond stable conditions can be necessary. This requirement for
high or ultra high pressure limits the application of these
materials due to production costs and the limited sizes and shapes
accessible with this technique.
[0009] On the other hand, infiltration has been successfully
utilised in generating fully dense composites even at low pressure
conditions. This is explained by the fact that even if/as pores are
generated within the structure during sintering, liquid phase is
continuously wicked up from the infiltrant source to fill these
pores. Effective infiltration therefore requires that the pores or
channels in the preform structure remain open for infiltration. The
limitation imposed by this pore size and density requirement means
that infiltration has been chiefly employed for the manufacture of
larger-grained diamond compacts, or those with a wide diamond grain
size distribution. Even under HpHT conditions (7.7 GPa,
1400-2000.degree. C.), infiltration of diamond powder with primary
grain size of .about.10 nm but secondary particle (agglomerate)
size of approximately 1 .mu.m was only possible to a depth of 2
mm.
[0010] This pore retention problem is exacerbated by the ongoing
formation of SiC within the preform. SiC formation from the
interaction of molten Si infiltrant and the carbon source is
accompanied by volume expansion of the solid phase. This reduces
the size of the existing pore channels and can result in blockage
thereof. This especially becomes a matter of concern for
fine-grained preforms, which already have an extremely fine pore
structure. An additional concern is that the formation of SiC is
strongly exothermic, which further accelerates the reaction in a
runaway effect.
[0011] Infiltration has a further advantage in that the purity of
the silicon source can be more adequately controlled through the
use, for example, of a monolithic silicon wafer. By contrast, a
reaction sintering or admixing technique typically requires that a
very fine powder be used in order to maximise microstructural
homogeneity. This brings with it the associated impurities of high
surface area particles, as well as concomitant contamination
introduced during the preparative mixing or milling process.
[0012] A further issue in the generation of diamond-SiC compacts
relates to the presence of free or elemental silicon in the final
binder phase. The thermal stability of a compact containing
discernible free silicon may be limited by the melting point of
silicon, as the bond between diamond and binder phase can be
compromised at this point. Typically the presence of free silicon
is the mark of an incomplete reaction with the carbon source. This
may occur where substantial SiC formation has masked or blocked off
the silicon melt from carbonaceous material, as diffusion of these
species through SiC is significantly slower than that along the
grain boundaries.
[0013] U.S. Pat. No. 4,124,401 describes a diamond compact
comprising a mass of diamond crystals adherently bonded together by
a silicon atom-containing binder. The compact is made by
infiltration under relatively mild hot pressing conditions (<1
kbar), where pressure is applied to dimensionally stabilise the
diamond mass before and during infiltration. The resultant binder
comprises SiC and a further carbide and/or silicide of a metal
component which forms a silicide with silicon. The diamond density
of the compact ranges from 70-90 volume %. The metal component for
the diamond body is selected from a wide group of metals such as
cobalt, chromium, iron, etc.
[0014] U.S. Pat. No. 4,151,686 describes a diamond compact similar
to that of U.S. Pat. No. 4,124,401 save that the resultant binder
comprises SiC and elemental or free silicon. The substantially
pore-free compact is generated at significantly higher pressures
(in excess of 25 kbar) through infiltration by an elemental silicon
melt. These high pressures are required in order to achieve the
characteristic high diamond density of the compact (from 80-95
volume %).
[0015] U.S. Pat. No. 4,664,705 discloses a method that infiltrates
a silicon alloy through a previously intergrown polycrystalline
diamond body, that was initially sintered in the presence of a
transition metal solvent/catalyst, where this previous binder has
been leached out. SiC forms in situ through the reaction of the
molten silicon with the intergrown diamond at HpHT.
[0016] U.S. Pat. No. 6,939,506 and U.S. Pat. No. 7,060,641 describe
the manufacture of fully dense diamond-SiC composites by reaction
sintering at HpHT conditions (namely 5 GPa and temperatures between
600-2000.degree. C.). The reagent mix is prepared by reactive
ball-milling of diamond powder (5-10 .mu.m particle size) and
crystalline silicon powder. At higher sintering temperatures, the
SiC binder that forms is nanocrystalline in nature; whilst at lower
temperatures residual unreacted elemental silicon tends to remain
in the binder phase. These compacts had a minimum possible
calculated diamond content of 77 mass %. It was observed that
ball-milling serves to transform the silicon to the amorphous
state, which was critical in determining the nanocrystalline nature
of the binder.
[0017] Another approach to the formation of SiC-diamond compacts is
disclosed in U.S. Pat. No. 5,010,043 and associated applications.
In a specific embodiment of this process, reaction sintering of a
diamond-silicon mixture is employed together with silicon melt
infiltration to form diamond-SiC compacts with a diamond density of
50-85 volume %. The silicon admixed within the compacts is
postulated to melt and wet the surfaces of the diamond particles,
establishing a continuous capillary system for infiltration. The
compact formation conditions are intermediate between conventional
HpHT and low pressure processes, at 10-40 kbar. Critical to this
process is a deliberate plastic deformation step that is observed
to significantly improve the properties of the resultant compacts
and enable the use of p and T conditions reduced from those of
HpHT. Given that it is known in the art that plastically deformed
diamond is inherently more reactive than diamond which is not (see
U.S. Pat. No. 6,680,914), it may be the case that the improved
reactivity of the diamond in this invention is what enables
effective bonding at lower p, T conditions. This is consistent with
the fact that manipulation of the sintering temperatures generates
compacts that contain minimal amounts of free silicon in the binder
phase, as the SiC formation reaction has been maximised.
[0018] It is also known in the art to produce diamond-SiC compacts
where the carbon source for the in situ SiC formation is not
dominantly supplied by crystalline diamond but by a carbon
introduced or produced on the diamond surface. Both low and higher
pressure techniques employing this approach are known.
[0019] U.S. Pat. Nos. 4,220,455 and 4,353,953 describe diamond-SiC
compacts formed by coating diamond particles with amorphous carbon
before infiltrating under partial vacuum with molten silicon. The
amorphous carbon is introduced by pyrolysis of organic binder
systems such as resins, polymers, etc., or by pyrolytic
decomposition of carbonaceous gases. An advantage of the resin or
polymer approach is that the organic residue can facilitate
formability of the pre-sintered diamond. It was additionally
observed that non-diamond carbon coatings were highly reactive in
the presence of molten silicon, easily wet by it and hence easily
formed SiC. However, the binder phase in these compacts still
comprised both SiC and unreacted elemental silicon.
[0020] U.S. Pat. No. 4,381,271 employs carbonaceous materials such
as fibrous graphite as an additional carbon source for SiC
formation. These fibres are admixed with coated diamond particles
before being infiltrated by molten silicon under a partial vacuum.
In the final compact binder both SiC and unreacted elemental
silicon were observed.
[0021] In most of these cases, any required pyrolysis is carried
out to minimise the graphitisation of the diamond; as this is seen
as detrimental to the potential properties of the compact. By
contrast, U.S. Pat. No. 6,447,852 and associated applications
disclose a low pressure infiltration process for the manufacture of
diamond-SiC compacts that utilises a deliberate graphitisation
step. Preferably 6-30 mass % of the diamond is deliberately
graphitised prior to infiltration with molten silicon. It is
postulated that the graphitised layer on the diamond surface
affects the pore character such that an optimal infiltration
environment results. A characteristic of compacts of this invention
is the discernible presence of free silicon in the binder
phase.
[0022] Infiltration remains a preferred method for the manufacture
of diamond-SiC compacts because of the opportunity it provides for
exploiting low pressure processes. There are significant cost
benefits inherent in this approach over using HpHT; and further
benefits of being able to access shapes and sizes not viably
attainable in HpHT or even medium pressure processes. However, the
use of infiltration for finer-grained diamond structures is
problematic because of the fine-scale nature of the pore structure
and the ease with which these pores can be blocked. Nonetheless,
finer-grained structures would be of great interest as high
performance composites. Additionally, the generation of a compact
containing no discernible free silicon that uses a low pressure
infiltration process would have significant cost and technical
benefits.
SUMMARY OF THE INVENTION
[0023] According to a first aspect to the present invention there
is provided an abrasive compact comprising a mass of diamond
particles and a silicon containing binder phase wherein the diamond
particles are present in an amount less than 75 volume % and the
binder phase contains less than 5 volume % unreacted (elemental)
silicon or silicide. Preferably the diamond particles are present
in an amount of more than 5, more preferably 10, and most
preferably 20 volume %; but less than 75, more preferably less than
70 volume %. The compact binder phase is characterised in that,
whilst it is dominated by a silicon-based chemistry, preferably
there is no detectable free or elemental silicon present in the
binder system and the majority of silicon present in the binder
phase is silicon carbide SiC. Preferably the SiC in the binder
phase is microcrystalline in nature. Preferably the diamond
particles are not plastically deformed to a significant degree and
the particles typically have an average grain size less than 10
.mu.m, more preferably less than 7 .mu.m and most preferably less
than 5 .mu.m. (Average grain size is measured using the largest
diameter of each grain or particle.)
[0024] Silicide results from the reaction of silicon with
impurities such as iron, etc.
[0025] Preferably the binder phase of the compact contains less
than about 4 volume % unreacted silicon, more preferably less than
about 3 volume % unreacted silicon, more preferably less than about
2 volume % unreacted silicon, most preferably less than about 1
volume % unreacted silicon.
[0026] Preferably the unreacted silicon content is within the range
of 0 to 5 volume %.
[0027] Still further according to the invention there is provided a
method of producing an abrasive compact including the steps of:
[0028] a. forming a feed diamond powder into a diamond preform,
[0029] b. interposing a separating mechanism between the diamond
preform and a silicon infiltrant source, [0030] c. heating the
diamond preform and the silicon infiltrant source until the
infiltrant is molten and the preform and infiltrant are isothermal,
and [0031] d. allowing infiltration from the molten silicon
infiltrant source to occur into the diamond preform.
[0032] Preferably the infiltration takes place with the application
of mild pressure (<1 kbar). More preferably infiltration takes
place while simultaneously removing the separating mechanism.
[0033] Preferably the feed diamond powder is coated with a
typically amorphous carbon layer through pyrolysis of an
appropriate organic binder. The compact may be a compact as
hereinbefore described.
[0034] SiC-diamond with low Si or other soft phases, preferably
none is suitable for armour applications (stopping high velocity
projectiles). As such, according to a third aspect to the present
invention there is provided armour comprising an abrasive compact
as hereinbefore described.
BRIEF DESCRIPTION OF THE DRAWINGS
[0035] FIG. 1 is a schematic set-up for the infiltration
experiments.
[0036] FIG. 2 is SEM micrographs diamond powders D2 and D9 showing
the effect of coating the diamond. The micrographs show the powder
before coating, fracture surfaces of the green compacts before
pyrolysis, and fracture surfaces of the preforms before
infiltration.
[0037] FIG. 3 is a graph showing average ratio of G-band intensity
to diamond Raman peak intensity for the D2 and D9 diamonds with
initial 5 and 20% resin after their pyrolysis.
[0038] FIG. 4A is a graph of pore size distribution in D2 diamond
preforms containing 5%, 10%, and 20% resin.
[0039] FIG. 4B is a graph of pore size distribution in D9 diamond
preforms containing 5%, 10%, and 20% resin.
[0040] FIG. 5A is a SEM micrograph showing the infiltration depths
of D2 containing 5% resin, after infiltration at 1500.degree. C.
for 30 minutes.
[0041] FIG. 5B is a SEM micrograph showing the infiltration depths
of D2 containing 10% resin, after infiltration at 1500.degree. C.
for 30 minutes.
[0042] FIG. 5C is a SEM micrograph showing the infiltration depths
of D2 containing 20% resin, after infiltration at 1500.degree. C.
for 30 minutes.
[0043] FIG. 5D is a SEM micrograph showing the infiltration depths
of D9 containing 5% resin, after infiltration at 1500.degree. C.
for 30 minutes.
[0044] FIG. 6A is a typical backscattered SEM micrograph of a
polished cross section of D17 containing 5% resin, after
infiltration at 1500.degree. C. for 30 minutes.
[0045] FIG. 6B is a typical backscattered SEM micrograph of a
polished cross section of D9 containing 10% resin, after
infiltration at 1500.degree. C. for 30 minutes.
[0046] FIG. 6C is a typical backscattered SEM micrograph of a
polished cross section of D2 containing 10% resin, after
infiltration at 1500.degree. C. for 30 minutes.
[0047] FIG. 6D is a typical backscattered SEM micrograph with
higher magnification of a polished cross section of D2 containing
10% resin, after infiltration at 1500.degree. C. for 30
minutes.
[0048] FIG. 7 is an example filtration process embodiment.
[0049] FIG. 8 is a graph showing pore size distributions for
diamond preforms.
DETAILED DESCRIPTION OF THE INVENTION
[0050] In the following description, reference will be made to the
following Figures:
[0051] FIG. 7 shows an example infiltration process embodiment,
and
[0052] FIG. 8 shows the pore size distributions for diamond
preforms made from a diamond powder with an average particle size
of 1.5 .mu.m with three different initial contents of phenolic
resin D2Pr05 shows the pore distribution at 5 mass % resin, D2Pr10
at 10 mass % and D2Pr20 at 20 mass %.
[0053] Compacts according to the present invention are typically
fine-grained diamond-SiC compacts (where the average diamond grain
size is typically less than 10 .mu.m) produced through infiltration
of a diamond preform by molten silicon-containing materials. These
compacts are unique in that they are free of detectable elemental
or free silicon in the final binder microstructure. Further, the
diamond in these compacts shows no significant plastic deformation.
The compacts of the invention further have a high relative diamond
density.
[0054] Compacts of the invention comprise a mass of diamond
particles distributed in a binder or binding phase. These diamond
particles will typically be uniformly distributed throughout the
binder phase. In order to achieve a suitable structure, it has been
found necessary for the diamond particles to be present in an
amount of more than 20, more preferably 30, and most preferably 40
volume %; but less than 75, more preferably less than 70 volume %
in the body. The diamond particles may be of natural or synthetic
origin. Diamond particles used in a preferred embodiment of this
invention have an average grain size less than 10 .mu.m, more
preferably less than 7 .mu.m and most preferably less than 5 .mu.m.
However, it is observed that many of the advantages of this
invention can also be realised where the diamond grain size is
coarser than in the preferred embodiment. The diamond particles may
have a monomodal, bimodal or multimodal size distribution.
[0055] The binder or bonding phase is dominated by a silicon-based
chemistry, however, there is less than 2% volume of detectable free
or elemental silicon or silicides present in the binder system of
the final compact and most preferably there is no detectable free
or elemental silicon present in the binder system of the final
compact. Typically the method used to detect free silicon is XRD
(X-ray diffraction). The binder typically comprises
microcrystalline SiC, although other silicon-based chemistries may
also occur. The silicon-based source for the infiltrant may be
elemental silicon or a suitable silicon alloy--if elemental
silicon, it may be in powder or monolithic form.
[0056] The compacts of the invention are manufactured using
temperatures that ensure that the infiltrant is molten, for example
in excess of the melting point of silicon (at approximately
1420.degree. C.): and extremely mild pressures less than 1 kbar.
Hence the manufacture process is characterised in that it occurs in
the thermodynamic region where diamond is metastable. These
conditions will be maintained for a time sufficient to produce the
abrasive body.
[0057] Preforms for the compacts according to this invention are
generated by initially coating the diamond feed diamond powder with
a suitable organic binder. In one embodiment of this invention
phenolic resin is used as the organic binder, although it will be
appreciated that other suitable binders may be used. Appropriate
levels for the initial coating are between 5 and 20 mass %, more
preferably about 10 mass %. The coated powder is then formed into a
green compact by cold compaction. The pore size and pore diameters
are controlled either by varying the compaction pressure on the
non-pyrolysed resin-diamond preform, or by varying the amount of
resin used. The green compact is then heat-treated to pyrolyse the
organic coating on the diamond powder compact under an inert
atmosphere (at temperature conditions where graphitisation of the
diamond will not occur). The green compacts generated by this
method retain sufficient structural integrity to be handled easily
and assembled into the infiltration assembly for subsequent
heat-treatment.
[0058] The preform is then infiltrated with molten silicon or a
silicon-containing alloy. The preform is placed into a suitable
reaction container in proximity to a silicon source, with an
appropriate separating mechanism being interposed between the
diamond preform and silicon source to space the diamond preform and
silicon source from each other. The container is heated to a
temperature in excess of the melting point of the silicon
(approximately 1420.degree. C.) or the silicon alloy; until the
diamond preform and silicon source are isothermal, and the silicon
source is molten. Gentle pressure (approximately 20 MPa) is then
applied in order to bring the preform and melt into physical
contact with one another and hence initiate infiltration.
Sufficient time is allowed for effective infiltration to occur and
then the container is optionally cooled.
[0059] The infiltrated compact is then removed from the container
and processed appropriately to achieve a suitable final
product.
[0060] The introduction of a suitable pyrolysed carbon layer onto
the surface of the diamond powder is preferable. Without being
limited by theory, it is assumed that the increased reactivity of
the amorphous carbon generated by the pyrolysis may allow rapid
initial SiC phase nucleation on the diamond/carbon surfaces during
initial infiltration. Counter-intuitively, this rapid nucleation
process appears to result in the formation of a controlled thin SiC
layer that effectively acts as a pseudo-barrier to the subsequent
diffusion of reactant species. Hence subsequent SiC growth can be
somewhat slowed and the potential runaway SiC formation which
results in pore blockage in fine-grained structures controlled. As
previously discussed, the carbon source in a similar low/no
pressure process (such as that disclosed in U.S. Pat. No. 6,447,852
and associated applications) arises from graphitic layers generated
in situ from deliberate graphitisation of the diamond powder. This
graphite layer, whilst more soluble and reactive than the diamond
itself is substantially less reactive than the amorphous carbon
layer of this invention. Hence the slower SiC formation in the
initial stages does not effectively mask the diamond surface and
prevent runaway SiC formation, leading to an increased probability
of pore blockage resulting in ineffective infiltration.
[0061] Also, the introduction of sacrificial non-diamond carbon
supplies the molten silicon with a non-diamond reactant, thus
sparing the valuable diamond phase from conversion into the softer
SiC one. Furthermore, and very importantly, the introduction of a
non-diamond carbon layer on the diamond particles results in an
increase of the pore size of the pyrolised diamond preform, as
shown in Figure B, thus providing the infiltrating silicon with an
easier passage. In the green compact, most of the carbon-supplying
resin occupies the pores of the diamond preform during initial
compaction. Therefore, the resulting non-diamond carbon that is
generated after pyrolysis is located in the diamond preform pores,
thus allowing for the diamond volume fraction to remain relatively
high while still supplying the advancing molten silicon front with
a carbon reactant.
[0062] Appropriate selection of the organic binder, required
additive levels and suitable pyrolysis cycle requires an
understanding of the yield and distribution of the amorphous carbon
layer that is generated. Whilst the preferred organic agent of this
invention is phenolic resin, it is anticipated that the use of
other similar organic materials would be self-evident to those
skilled in the art such as paraffin, polysaccharides acrylates,
etc. The organic binder is additionally useful in that it allows
the generation of a pressed green compact that has some strength
i.e. can be freely handled and machined. The organic binder of the
preferred embodiment is typically introduced into the diamond
powder mix in dissolved form in a suitable organic solvent such as
acetone. Alternative solution methods such as spraying, or gaseous
techniques such as the in situ decomposition of a natural gas on
the diamond surface would equally be obvious to those skilled in
the art.
[0063] Unfortunately, the engineered increased reactivity of the
coated fine diamond was observed to result in a premature reaction
in the contact region between the preform surface and the silicon
infiltrant, whilst the latter was still in the solid state during
the heating cycle. This reaction was seen as highly undesirable
because the early generation of SiC in this region would easily
block the very fine pore structure of a fine-grained diamond
preform, resulting itself in incomplete infiltration. This
phenomenon was further exacerbated by the increased viscosity of
the infiltrant during the early stages of infiltration before it
was fully molten. Any drop in temperature from the infiltrant
source to the diamond preform was also found to be extremely
disadvantageous, as cooling of the infiltrant within the preform
had a similar disruptive effect.
[0064] The identified problem was therefore to prevent a premature
reaction at the interface between the diamond preform and silicon
source whilst it was still in the solid state; and to ensure that
the diamond preform and molten silicon source were isothermal
before they were brought in contact. Any separation mechanism
additionally required the facility to be triggered remotely in situ
during the sintering cycle.
[0065] A set of SiC, SiC-based ceramic foam or graphite felt
spacers (stilts) was designed to fit into the interface region
between the silicon source and diamond preform. The dimensions of
these spacers were chosen such that they did not create a physical
barrier per se between the two parts, but interposed a space
between them. Hence and by way of example, in the case of an 18 mm
diameter preform, three SiC spacers of approximately 2 mm.times.2
mm.times.3 mm were used to separate the preform and silicon source.
These spacers functioned as effective stilts, maintaining
separation between the two parts until, once the silicon source was
molten; the application of external pressure forced them down into
the molten silicon source and allowed contact. The "stilt" spacers
must be of such a material that they remain solid during the course
of the reaction and are chemically inert with respect to the
infiltration reaction. In addition to the above the "stilts" can
also be silicon-infiltrated silicon carbide or recrystallised
silicon.
[0066] The combined effect of the pyrolytic carbon layer in
increasing reactivity, coupled with a pore maintenance; and the
physical separation of the infiltrant and preform until
infiltration conditions are optimal, allows diamond-SiC compacts
with various unique characteristics, namely: [0067] the elimination
of free silicon in the binder phase [0068] the effective
infiltration of finer-grained diamond preforms [0069] increased
diamond density over that achieved with known low pressure
infiltration routes due to the use of a non-diamond source for at
least a part of the SiC formation.
[0070] Essentially, when the diamond content in a compact is high,
the likelihood and content of elemental Si being present in the
finished article is greatly reduced for the following reasons:
[0071] Where the diamond content is high, and especially where the
grains are fine, higher pressures are typically required in order
to compact the material sufficiently and drive infiltration. Higher
pressures may have the benefit of driving the diffusion of Si and C
and promoting the reaction to form SiC; [0072] Where the diamond
content is high, the pores may typically be relatively smaller,
resulting in smaller isolated volumes of unreacted, free Si.
[0073] The present invention teaches low or no Si even where the
diamond concentration is relatively low and/or the diamond is
relatively fine.
[0074] The invention is further illustrated by the following
non-limiting examples:
Example 1
[0075] A preform containing diamond powder (average grain size of
1.5 .mu.m) coated with a pyrolytic carbon layer was prepared.
[0076] An amount of phenolic resin to give 10 mass % in the diamond
mix, was dissolved in acetone at a concentration of approximately
34.3 g/l. This solution was then mixed with the diamond powder and
heated in a water bath to 70-80.degree. C., whilst stirring, to
evaporate off the acetone. The resulting agglomerated powder was
crushed and screened using a .about.325 mesh screen. SEM
micrographs of the coated grit showed that the resin was
homogeneously distributed on the diamond surfaces, both before and
after pyrolysis.
[0077] A green compact was then formed by cold compaction of the
screened powder at ca. 60 MPa. This green compact was then
heat-treated at 120.degree. C. in air for 18 hours, in order to
cure the resin. The resin coating on the diamond was then pyrolysed
by heat treatment under argon. The heating upramp cycle was in two
parts: initially up to 450.degree. C. at 2.degree. C./min; followed
by heating to 750.degree. C. at 10.degree. C./min. The preform was
then held at 750.degree. C. for 1 hour. After cooling, the porosity
of the preform was determined to be approximately 30%. From the
weight loss it was evident that about half the mass of the resin
had volatilised and left the compact.
[0078] The preform was then infiltrated with molten silicon under
very mild pressure.
[0079] A silicon infiltrant source body 5 was placed inside an
hBN-coated graphite pot 2 such as that shown in FIG. 7. Three SiC
separating spacers 4 (of dimension such that they served a "stilt"
function as previously discussed) were placed on top of this source
5. The diamond preform 3 was then placed in the pot 2. An
hBN-coated graphite piston 1 was then inserted into the pot 2. The
pot 2 was heated to 1500.degree. C. at a rate of 50.degree. C./min.
Once the temperature inside the container reasonably exceeded the
melting point of silicon (.+-.1420.degree. C.), a pressure of 20
MPa was applied to the piston 1. This brought the preform 3 and
molten infiltrant 5 into contact, commencing the infiltration
process. The temperature was held at 1500.degree. C. for
approximately 30 minutes before cooling. (Pressure was continued
even during the cooling cycle until the temperature reached
1300.degree. C.).
[0080] The infiltrated sample was recovered from the pot and
investigated. Microstructural analysis showed that the compact was
well infiltrated to a depth of at least 2.5 mm. The infiltrated
volume was observed to be completely free of pores, with a high
concentration of diamond. XRD analysis showed only diamond and SiC,
with no residual unreacted elemental or free silicon present in the
compact. The diamond content of the compact was estimated to be
approximately 40 volume %, with the remainder being SiC phase.
Examples 2-7
[0081] Further diamond compacts was prepared according to the
method of example 1, save that the diamond average grain size and
phenolic resin content were altered as shown in Table A.
[0082] Table A Summaries of various characteristics of the compacts
produced.
TABLE-US-00001 TABLE A Diamond Preform grain resin Infiltration
Phase composition size content depth (volume %) Example (.mu.m)
(mass %) (mm) Diamond SiC Si 1 1.5 10 2.5 40 60 0 2 9 10 full 53 47
0 3 1.5 5 1.25 -- -- -- 4 9 5 2 46 51 3 5 1.5 20 poor -- -- -- 6 9
20 poor -- -- -- 7 16.5 5 full 52 40 8
[0083] As is evident from Table A, excess quantities of phenolic
resin are undesirable in that they cause a similar pore-blocking
effect to that observed without any resin being present. In this
case, optimal levels of resin addition at approximately 10 mass %
were observed to maximise the infiltration process and reduce the
presence of undesirable free silicon.
Example 8
[0084] The contents of the paper `The low-pressure infiltration of
diamond by silicon to form diamond-silicon carbide composites` as
authored by Sigalas, Herrmann and Mlungwane is incorporated herein
by reference. For the avoidance of doubt, the paper is set out
below:
Abstract
[0085] The infiltration of fine-grained diamond preforms by molten
silicon is limited by the blocking of the pores as a result of the
volume increase during the reaction of diamond with SiC. Therefore
in the present paper the infiltration of preforms made with diamond
powders with different grain sizes was investigated. The preforms
were prepared using phenolic resin as a binder. With increasing
resin content the pore size increases, but the pore volume
decreases. As a result the infiltration depth increases strongly
for medium resin content. For the fine-grained .about.1.5 .mu.m
diamond preforms, a maximum infiltration depth of 2.5 mm is
obtained at 10% resin, whereas at 5% resin only 1.25 mm could be
infiltrated.
1. INTRODUCTION
[0086] Diamond is the hardest material known to man. Because of
this, it finds extensive industrial application where ultra-hard
material properties are needed. Due to its high hardness, it is
difficult to make diamond tools of different shapes and sizes
purely from cutting and shaping diamond. This has led to the
development of diamond composite materials which consist of small
diamond grains either sintered together through a liquid phase
sintering process, or held together in a matrix by a binder phase
material. The former process gives rise to the class of
polycrystalline diamond materials (PCD), while the latter results
in a number of composite materials, of which the foremost is that
of SiC-diamond composites. The introduction of the second phase
improves the formability and the fracture toughness of such
diamond-based materials.sup.1.
[0087] Metallic phases such as cobalt are present in PCD and are
commonly used as liquid phase sintering aids in the production of
that material. These metals however were found to catalyze the
graphitization of diamond thus limiting the application
temperatures of these PCD materials to below 1000.degree. C..sup.1.
Silicon carbide has been found to be exceptionally good as a
diamond binder phase. Because of the structural similarities
between diamond and silicon carbide, a strong bond forms between
them.sup.2 resulting in a material with a very strong adhesion
between the diamond grains and the SiC matrix. Silicon carbide does
not react with diamond and the composite material can be used at
temperatures above 1000.degree. C. Application temperature is
limited by the melting temperature of silicon if some unreacted
silicon is present in the final product.
[0088] SiC is commonly formed in situ from a reaction between
diamond and/or amorphous carbon or graphite with silicon. The
silicon can be introduced into the diamond in different ways,
either by infiltrating molten silicon into a diamond preform or by
reaction sintering silicon powder and diamond powder.sup.3, 4,
5.
[0089] The main production route of these composites includes the
use of high-pressure and high-temperature in order to achieve
sintering within the regions of diamond stability [6]. Use of high
pressures however restricts the range of applications of these
materials due to high cost of production and the limited range of
possible sizes and shapes of the products made. Some attempts.sup.5
have been made to produce this composite material under conditions
of low pressure (i.e. in the diamond metastable region). Hot
Isostatic Pressing (HIP) method was employed at a maximum pressure
applied of 20 MPa. A product more than 90% dense was obtained. It
is of great importance to note that for the reaction sintering
route, if the reaction proceeds under low pressure conditions,
voids are produced within the body because of the volume reduction
occurring during the reaction.sup.7.
[0090] The advantage of infiltration as stated by J. Qian et
al.sup.7, is that the liquid phase keeps filling the pores in the
diamond skeleton and hence a more dense material is produced.
Infiltration can also be successfully performed at low pressures
giving a dense product.
[0091] Infiltration on the other hand has been successful under low
pressure conditions only for large grained diamond preforms (7-63
.mu.m grain size).sup.3, 4. It should be noted that in these
materials a wide grain size distribution was used. Even under high
pressure (7.7 GPa, 1400-2000.degree. C.), E. A. Ekimov et al..sup.8
could infiltrate diamond powder with primary grain size of
.about.10 nm but secondary particle (agglomerate) size of .about.1
.mu.m only up to an infiltration depth of 2 mm.
[0092] Therefore the aim of this study is to investigate the
infiltration of diamond by silicon using minimal pressure, and to
analyze the limitations accompanying the infiltration of small
diamond grain size preforms.
2. EXPERIMENTAL
2.1. Preform Preparation
[0093] Preforms were produced using three different diamond
powders, labelled D2, D9 and D17 (Element Six (Pty) Ltd). The
characteristics of these powders are given in Table 1. The
composition of the diamond preforms was modified by the addition of
phenolic resin (Plyophen 602N; Fa. PRP Resin). This component was
necessary for the formation of the preform during pressing. It acts
as a lubricant and a binder. Resin concentrations of 5, 10 and 20
wt % were investigated. The composition and names of the samples
are given in Table 2. For the preparation of the preforms phenolic
resin was dissolved in acetone (34 3 g/l) and mixed with the
diamond powder. This suspension was stirred continuously while kept
in a water bath at 70-80.degree. C. to evaporate off the acetone.
The resulting powder is agglomerated the degree of agglomeration
increasing with increasing resin content and decreasing diamond
particle size. The agglomerated powder is crushed and screened
using a .about.325 mesh screen. The screened powder is pressed into
a green compact of 18 mm diameter and 5 mm height under 60 MPa of
pressure for about 5 seconds.
[0094] The green compacts were heat treated at 120.degree. C. for
18 hours to cure the resin in air. They were then weighed and the
resin pyrolysed under argon by heating at a rate of 2.degree.
C./min up to 450.degree. C. followed by 10.degree. C./min up to
750.degree. C. where a dwelling time of 60 minutes was undergone.
Cooling to room temperature was carried out at a rate of 10.degree.
C./min.
[0095] The preforms' green density and porosity were determined
after pyrolysis. The green densities were calculated from the mass
and volumes of the preforms while the porosity and the pore size
distributions were determined using a mercury porosimeter
(Quantachrome Poremaster-60). Raman spectra were acquired with a
Jobin-Yvon T64000 Raman spectrometer operating in a single
spectrograph mode with an 1800 lines/mm grating. These measurements
were performed in order to determine the uniformity of the resin
coating. For each sample a line 1000 micron in length and
consisting of 100 points was mapped in the central region of the
sample using a motorized XY stage.
2.2. Infiltration
[0096] An excess amount of silicon powder (1-20 .mu.m Goodfellow)
was cold pressed into an 18 mm diameter tablet. This tablet is then
placed in an hBN-coated graphite pot (FIG. 1). The diamond preform
is placed on top of this Si tablet. Three SiC pieces of
2.times.2.times.3 mm size are used to separate these two tablets so
that no reaction in the solid state, during heating up, can take
place. An hBN-coated graphite piston covers the pot. The set-up was
heated up at 50.degree. C./min to 1500.degree. C. at which
temperature it dwelled for 30 minutes. Cooling was achieved at a
rate of 20.degree. C./min. Pressure (20 MPa) is applied onto the
piston after the temperature exceeds that at which silicon melts
(.+-.1420.degree. C.) to bring the preform and the melt into
contact so that infiltration can commence. It is then released when
the temperature reaches 1300.degree. C. during cooling.
[0097] The products of the infiltration were cross-sectioned. The
cross sections were polished using resin bonded diamond wheels with
1 .mu.m diamond at 3000 rpm before characterization with SEM and
XRD.
[0098] The phase composition of the infiltrated materials was
determined by quantitative image analysis using Image Tool3.
3. RESULTS
3.1 Preforms
[0099] SEM micrographs of two of the diamond powders used and the
powders mixed with the resin are shown in FIG. 2. It can be
inferred that the resin coated the diamond homogeneously both
before and after pyrolysis. This was confirmed also by the Raman
spectroscopy measurements. FIG. 3 indicates that both the materials
produced from D2 and D9 which initially had 20% resin have thicker
graphitic carbon layers than their 5% counterparts. The main
graphitic carbon G-band gave fairly constant peak intensity in all
samples for all the mapped points, indicating fairly uniform
coverage by the resin.
[0100] The pore size distribution determined by Hg-porosimetry is
given in FIG. 4A and FIG. 4B for the preforms prepared from diamond
powder D2 and D9. In Table 2 the green densities and mean pore
channel diameter are given. An increase in the resin content
increases the average pore diameter while decreasing the pore
volume. The decrease of the pore volume is more pronounced for the
smaller diamond grain sizes. Nevertheless the overall green density
is nearly constant.
3.2 Infiltration Results
[0101] The results of the infiltration experiments for the
different preforms are given in table 2. The micrographs in FIG.
5A, FIG. 5B, FIG. 5C, and FIG. 5D show the cross sections of
infiltrated samples. The infiltration depth for the different
materials is clearly visible. Increasing the amount of the resin in
the preforms up to 10 wt % improves the infiltration of the green
compacts for the materials produced from the low grain sizes
diamonds, e.g. for the material D2Pr05 with 5% resin the
infiltration depth was only 1250 .mu.m and increases up to 2500
.mu.m for the material with 10 wt % resin (D2Pr10).
[0102] The SEM micrographs of the polished sections in FIG. 6A,
FIG. 6B, FIG. 6C, and FIG. 6D clearly indicate that the infiltrated
areas are completely free of pores and with a high concentration of
diamond. This could be confirmed by XRD. The black phase is
diamond, the white (where present) is free silicon and the grey
phase is SiC. While in the coarse grained product the presence of
free silicon is obvious (the white phase), this is not detectable
for the materials with the medium and fine diamond powders, where
one can only see the black diamond phase and the grey SiC phase.
The amount of diamond determined by image analysis could be
slightly overestimated.
4. DISCUSSION
[0103] As was shown previously.sup.9 diamond is well wetted by
liquid silicon at temperatures higher than 1450.degree. C.
Therefore a pressureless infiltration would be possible.
[0104] The infiltration is hindered by the formation of SiC surface
layers on the diamond, which can block the pore channels and reduce
the infiltration depth. Additionally the silicon will react with
the added phenolic resign. The investigations of the reaction of
liquid silicon with CVD-diamonds, glassy carbon and graphite has
shown.sup.9-11 that the reaction in all cases results in a very
fast formation of protective SiC-layers with similar thickness. The
reaction is faster for less crystalline carbon sources, in the
infiltrated samples no residual non diamond carbon was observed.
This indicates that the resin converts preferentially into SiC.
[0105] The fast reaction of the carbon with liquid silicon results
in blocking of the pore channels and is also the reason why
infiltration experiments so far were successful only with preforms
made of diamonds having large pore sizes.sup.3-4. The reaction of
silicon with diamond or other carbon sources is further enhanced by
the strong exothermic character of the interaction of silicon with
carbon. This results in a pronounced heat up of the system.sup.10
and an acceleration of the reaction resulting in premature blocking
of the pores.
[0106] The pyrolysed resin in the sample strongly changes the
microstructure of preforms. It increases the pore channel diameter,
e.g. by a factor of 1.5 times for D2Pr samples and by a factor of 3
for the samples with the medium grain size (D9).
[0107] Unfortunately at the constant pressure used for the
preparation of the preforms the pore volume decreases with
increasing resin content, i.e. pores between the diamond particles
are filled by the pyrolysed resin. The reduction of the pore volume
is more pronounced for the low grain size diamond composites
(nearly 70%) whereas the change for the samples made with D9 powder
it is only 38%. This reduction can be reduced by decreasing the
pressure during compaction of the preform.
[0108] Small amounts of resin (5 wt %) are needed to make the
pressing of the diamond powder possible. Without the presence of
resin no pressed samples could be prepared. The resin coats the
diamond particles (FIG. 2). This coating plastically deforms during
pressing and glues the diamond particles together. With increasing
resin content the resin will begin to fill the pores of the
preform, during the pressing process, starting with the smaller
ones. Therefore only the larger pores will remain and the overall
porosity will be reduced. If the diamond particles had a constant
packing density in the green body and the resin fills only the
pores then the green density had to be increased with increasing
resin content. In the investigated samples the density reduces
slightly with the increasing resin content. This indicates that the
distance between the diamond particles increases with increasing
resin content.
[0109] To some extend the pore structure in the high resin content
materials can be related also to the structure of the granulates
prior to the pressing of the preform. However no inhomogeneity of
the diamond, Si and SiC distribution was found after infiltration
(FIG. 6A, FIG. 6B, FIG. 6C, and FIG. 6D).
[0110] This changed pore structure with increasing resin content
will influence infiltration in the following ways: [0111] The
increase of the pore channel radius will improve the infiltration.
[0112] Therefore for preforms with up to 10 wt % resin content a
strong increase in the infiltration depth was observed. [0113] The
reduction of the overall porosity by deposition of the resin
between the diamond particles will reduce the infiltration depth
due to the possibility of blocking the pores. The volume increase
during the reaction of diamond with liquid silicon is much larger
than for the amorphous carbon or graphite with silicon. Therefore
the reaction of the resin with liquid silicon will result in the
blocking of the pores to a lesser extend. This will reduce the
influence of the reduction in porosity. It was shown, that carbon
preforms with overall densities less than 0.9 g/cm.sup.3 can be
fully converted into SiC.sup.12. Therefore the resin themselves
with a density of less than 1 g/cm.sup.3 can be converted
completely. Therefore the medium resin content improves the
infiltration and only high resin content decrease the infiltration
due to the lower porosity. Therefore the reduction of the porosity
has only a decisive influence on the infiltration at higher resin
content (20%).
[0114] The study of the interaction of diamond with molten silicon
has shown that after the onset of the interaction, a SiC layer of
5-10 .mu.m thickness is formed very quickly on the surface of the
diamond particles. The thickness of the layer is controlled by the
density of the nuclei formed. If the amount of nuclei is large the
thickness of the layer directly formed would be lower [9] and
infiltration would be possible to a higher infiltration depth. A
similar effect could be caused by the faster reaction of the
pyrolysed resin, which would help improve the infiltration
additionally. For the material D22Pr5 after 30 min infiltration the
thickness of the SiC-layers formed on the diamonds can be estimated
to be in the range of 2-5 .mu.m (FIG. 6A). This value is less than
what was observed in model experiment with CVD-diamond
plates.sup.9.
[0115] The resin has the additional effect that a smaller amount of
diamond is converted to SiC. Therefore high amounts of diamond were
observed in our samples after infiltration.
[0116] The large grained products contain some free silicon due to
their large pores in the preforms. The Si, which remains after
formation of the dense SiC layer around the diamond, reacts only
very slowly because this reaction is controlled by the diffusion
through the SiC-layer.sup.9-11. The medium and fine grained
products have no detectable free silicon in them which is in
agreement with this explanation.
5. CONCLUSION
[0117] The investigation of the infiltration of diamond preforms
produced from mixtures of phenolic resin and diamond of different
grain sizes from 1.5-17 .mu.m can be summarized as follows: [0118]
1) The addition of the resin allows a simple shaping of preforms.
[0119] 2) Increasing the amount of resin causes pronounced
increases of the pore channel diameter and reduces the amount of
porosity at similar green densities, because the resin fills
partially the space in between the skeleton formed by the diamond
particles. [0120] 3) Despite the fact that the overall porosity is
reduced by adding the resin, the infiltration depth increases by a
factor of two for the D2Pr10 in comparison to the D2Pr05. Similar
effects were found for the samples with coarser grain size
(D9Pr10). [0121] 4) For a larger resin content the infiltration
depth decreases again strongly due to the much lower pore
volume.
REFERENCES
[0122] The following references are included herein by reference
[0123] 1 Tomlinson P. N., Pipkin N. J., Lammer A. and Burnand R. P.
Indust. High performance drilling-Syndax 3 shows versatility.
Diamond Rev., 6 (1985) 299. [0124] 2 Qian J., Voronin G., Zerda T.
W., He D. and Zhao Y. High-pressure, high-temperature sintering of
diamond-SiC composites by ball-milled diamond-Si mixtures. J. Mater
Res., 17 (8) (2002) 2153. [0125] 3 Gordeev S. K., Danchukova L. V.,
Ekstroem T., Zhukov S. G. Method of manufacturing a diamond-silicon
carbide composite and a composite produced by this method.
CA2301775, 1999. [0126] 4 Gordeev S. K., Zhukov S. G., Danchukova
L. V., Ekstrom T. Method of manufacturing a diamond-silicon
carbide-silicon composite and a composite produced by this method.
EP1253123, 2002. [0127] 5 Shimono M. and Kume S., HIP-Sintered
Composites of C (Diamond)/SiC. J. Am. Ceram. Soc., 87 (4) (2004)
752. [0128] 6 Hall H. T. A Synthetic Carbonado. Science I, 169 (39)
(1970) 865. [0129] 7 Hillig W. B., Making ceramic composites by
Melt Infiltration. American Ceramic Society Bulletin, 73 (4) (1994)
56. [0130] 8 Ekimov E. A., Gavriliuk A. G., Palosz B., Gierlotka
S., Dluzewski P., Tatianin E., Kluev Y., Naletov A. M. and Presz A.
High-pressure, high-temperature synthesis of SiC-diamond
nanocrystalline ceramics, App. Phys. Lett., 77 (2000) 954. [0131] 9
Mlungwane K., Sigalas I., Herrmann M. and Rodriguez M., The wetting
behaviour and reaction kinetics in diamond0silicon carbide system.
Submitted for publication In Diamond and Related Materials. [0132]
10 Sangsuwan P., Tewari S. N., Gatica J. E., Singh R. N., and
Dickerson R. Reactive Infiltration of Silicon Melt through
Microporous Amorphous Carbon Preforms, Metallurgical and Materials
Transactions B, 30B (1999) 933. [0133] 11 Zhou h., Singh R. N.,
Kinetics model for the growth of silicon carbide by the Reaction of
Liquid Silicon with Carbon. Journal of the American Ceramic Society
78 (9) (1995), 2456-2462. [0134] 12 Siegel S., Petasch u., Boden
G., Biogene Keramik--eine Alternative?. Keramische Zeitschrift,
(2004), 4 234-238.
TABLE-US-00002 [0134] TABLE 1 The mean particle size of the three
diamond grades used in the experiments. Mean particle size (.mu.M)
Diamond grade D (v, 0.5) D (v, 0.9) D2 1.51 2.46 D9 9.02 16.42 D17
16.82 22.38
TABLE-US-00003 TABLE 2 A summary of the infiltration results of the
performs containing different amounts of resin. GREEN WEIGHT
DENSITY MEAN PHASE RESIN LOSS (AFTER PORE INFILTRATION COMPOSITION,
DIAMOND CONTENT, DURING PYROLYSIS), POROSITY, DIAMETER, HEIGHT, VOL
% SAMPLE POWDER wt % PYROLYSIS, % g/cm.sup.3 % .mu.m .mu.m Diamond
SiC Si D2PR05 D2 5 1.86 .+-. 0.03 1.82 40 0.47 1250 0 D2PR10 10
4.11 .+-. 0.01 1.80 29 0.59 2500 0 D2PR20 20 8.73 .+-. 0.07 1.79 11
0.77 17 36 64 0 D9PR05 D9 5 2.00 .+-. 0.02 1.84 38 2.7 2000 46 51 3
D9PR10 10 4.10 .+-. 0.02 1.78 29 4.9 5000.sup.1) 53 47 0 D9PR20 20
9.00 .+-. 0.03 1.71 15 8.8 97 0 D17PR05 D17 5 2.2 .+-. 0.02 1.97 25
5.7 5000.sup.1) 52 40 8 D17Pr10 10 2.17 30.2 6.8 5000.sup.1)
.sup.1)Fully infiltrated
* * * * *