Keywords to match content
Learn to Google Bomb Immortal? Taking Over Internet Taking Over Search Diamond MicroChips
Aren't you just fucking amazed?Vbull,

Diamond microcrystals have been synthesized using ultrasonic cavitation of a suspension of hexagonal graphite in various organic liquid
media, at an average bulk temperature of the liquid up to 120°C and at atmospheric pressure. The yield of diamond is up to 10% by mass. The
diamonds were characterized by scanning electron microscopy, X-ray diffraction and laser Raman spectroscopy. Analysis of the crystallite size
distribution showed that the diamonds were nearly mono-dispersed, having a size 6 or 9μm ± 0.5μm, with cubic, crystalline morphology.
© 2008 Elsevier B.V. All rights reserved.
Keywords: Ultrasound cavitation; Microcrystalline diamond
1. Introduction
Synthesis of diamond by a variety of chemical and physical
routes is currently a topic of great scientific and commercial
interest. Production of single crystal diamond particles of sizes
between a few nm and several mm usually require process
pressures in excess of tens of thousands of atmospheres and
temperatures N 2000K in order to convert graphite into
diamond. This is the basis of the so-called high pressure–high
temperature (HPHT) method, which produces much of the
world's supply of ‘industrial diamond’ used for commercial
cutting and drilling applications.
An alternative strategy for generating these high tempera-
tures and pressures was reported by Flynn in 1986 [1]. He
suggested that ultrasonic cavitation processes, if sufficiently
powerful, should produce the necessary combination of
pressure and temperature to allow graphite-to-diamond trans-
formation in metal melts. Cavitation is an efficient method to
concentrate low density elastic wave energy into higher
densities, as a result of the rapid collapse of cavitation bubbles
produced in a suitable liquid medium [2]. The overall picture of
cavitation bubble formation is as follows. As an elastic (sound)
wave passes through a liquid, it produces alternating regions of
reduced density (negative pressure) and increased density
(positive pressure). If the sound wave is sufficiently intense,
the reduced density regions form cavities (bubbles) filled with
the saturated vapor of the liquid. Any gases dissolved in the
fluid diffuse through the cavity walls and also contribute to the
vapor inside the bubble.
Inthecontraction phase,thecavitycollapses under the effectof
positive pressure and surface tension forces, and the vapor-gas
mixture within experiences a rapid, strong adiabatic compression.
Depending upon the cavitation conditions, at the moment of
collapse, the pressure, p, and temperature, T, inside the cavity
may reach extremely high instantaneous values, p ∼ 10
T ∼ 1000K [3]. As a result of the simultaneous collapse of many
cavities, a cavitation zone is formed in the ambient fluid. The
cavitation zone can be considered to be a peculiar kind of “power
transformer”, in which energy is accumulated rather slowly
(∼ 10ms) during the negative pressure phase, but which is
released on a very short timescale (∼ 1ns). As a result, the
Available online at
Diamond & Related Materials 17 (2008) 931–936
⁎ Corresponding author.
E-mail address: (S.G. Aloyan).
0925-9635/$ - see front matter © 2008 Elsevier B.V. All rights reserved.
instantaneous power is many orders of magnitude greater than
that of the average power introduced to the cavitation medium.
These extreme conditions create a specific physical and chemical
medium for realization of many chemical reactions, such as for
obtaining nanomaterials [4–6]. The instantaneous high pressure–
high temperature conditions also provide the correct environment
for the graphite-to-diamond transition.
In 1974, Galimov published a theoretical study concerning the
possibility of synthesising diamonds using the cavitation
phenomenon [7]. Using a prototype ultrasonic reactor, he and
co-workers then corroborated the prediction by successfully
synthesising nanodiamonds by cavitation destruction of bensole
[8]. However, the approach (nucleation of diamond directly from
cavitation of liquid hydrocarbons) and the apparatus were low
power and very small scale. Although they proved the concept
was feasible, the yield (and efficiency) for their process was too
low for it to be a commercial possibility. The most significant
work on the development of cavitation methods for nanodiamond
synthesis is the US patent “Method and means for converting
graphite to diamond” granted to Hugh G. Flynn in 1986 [1]. The
author discussed practically all aspects of cavitation processes in
the melts of easily fusible metals (Al, Ga, In, Sn). The device
construction is presented, and the conditions for achieving
maximal cavitation effect on solid (graphite)-fluid (metal melt)
boundary are described. A scheme of maintenance of spherical
symmetry for the cavitation pocket is proposed to achieve
maximum cavitationimpulse.Althoughthe apparatus and various
technological aspects of the synthesis process are discussed in
detail, there is no subsequent description of the synthesis product.
We must therefore conclude that Flynn never successfully
synthesised diamond by this route.
Other attempts at diamond synthesis by cavitation include
Leonov etal. [9], whoreportedultrasoniccavitation within hexane
and ethanol, although the nanoparticles they produced were not
characterized and may have been graphitic. Wang et al. [10] and
Pearce et al. [11] energized cavitation by pulsed laser ablation
(PLA) onsolid(graphite) atafluid(water,cyclohexane)boundary.
However, diamond synthesis under PLA conditions is not energy
efficient, and only nanosized diamond particles with low yield
were reported using this method. For nanodiamond synthesis,
generation of a high power cavitation impulse is not necessary.
Therefore, it is possible to use practically any cavitation media,
from organic (high- and low-boiling) to inorganic fluids (H
[11]. For production of larger diamond crystallites (μm or even
mm in size) cavitation effects in the presence of a solid carbon
phase, such as a graphite target or suspended graphite particles
may be the way forward.
During the last ten years, theoretical and experimental research
into the growth dynamics and collapse of cavitation bubbles has
achieved pressures (p ∼ 10
bar) and temperatures (T
7000K), sufficient for thermonuclear reaction in deuterium/
acetone [12–15]. Authors of these works have applied a new
approachtovapourcavitationinorder toovercomethe problem of
the cavitation bubbles losing their spherical symmetry during
collapse — previously a major constraining factor for the
maximum temperatures and pressures attainable. The process of
cavitation bubble collapse was studied by Nigmatulin et al.
[15,16]. Their calculations were based on the HYDRO program
developed by Moss et al. [17] with a series of additions, which
allowed the theory to be extended to higher temperatures and
pressures. The description of the bubble collapse was based on
numerical solution of a set of equations in the form of mass,
energy and impulse conservation laws for gaseous and fluid
phases, assuming spherical symmetry of the collapsing cavity,
and taking into account the medium velocity, pressure and
temperature. They included also the Hertz–Knudssen–Lengmure
conditions on the phase boundary, taking into account the fluid
condensing coefficient, α.
The results of these investigations highlighted the importance
of the cavity medium properties for the formation of super-high
pressure shockwaves when the cavities collapse, and determined
the cavitation fluid parameters at which the spherical shape of the
bubbles is maintained up to the final stage of collapsing. These
include that the cavitation medium should be a well degassed
(organic) fluid composed of large molecules, with a low vapour
pressure, and have a low vapour-phase acoustic velocity, a high
condensation coefficient (α ∼ 1) and high cavitation strength.
For the right choice of medium, cavitation could be an ideal
method for generation of a high-power cavitation impulse. In
these conditions, the temperature drop behind the pressure-
shockfront occurs extremely rapidly, ∼ 10
− 1
cavitation processes, ultra-high energies are possible for time-
scales of a few ns, which can act locally upon the substances in the
region of the collapsing cavities resulting in shock compressions.
Within these compressions, strongly developed density fluctua-
tions allow rapid and reversible clustering of substances, where
the state of the substance (solid, liquid or vapour) can vary rapidly
[18,19,]. This is exactly what is required for the extreme
compression of graphite needed for diamond synthesis [20,21].
The purpose of this investigation is to demonstrate the
feasibility of using such high power cavitation processes to
produce microsized synthetic diamond particles.
2. Experimental
Fig. 1 shows the cavitation module designed for microdia-
mond synthesis. The module consists of a reactor (1), composed
Fig. 1. The experimental cavitation module used in this study.
A.K. Khachatryan et al. / Diamond & Related Materials 17 (2008) 931936
of a 100mm-long tube made from stainless steel or heat-resistant
glass with inner diameter of 25mm. The central portion of the
reactor has a charging hole (2), from which to input or extract the
liquid media. Two ultrasonic emitters (3), attached to a package of
titanium concentrators with ZTBS-3 piezoceramic discs (4), are
fixed to opposite ends of the reactor. The radiating surface
diameter is 20mm for each concentrator. The reactor (1) is water
cooled and controlled by a thermostabilization feedback system
(5). If required, the emitters can be cooled by radiators (6).
A 23kHz sine-wave generator with output power of 1000W
and self-tuning resonance frequency was used to initiate
cavitation. The maximum ultrasound intensity in the reactor
centre was 75–80W cm
− 2
(corresponding to a sound pressure
amplitude of about 15–16bar). The ultrasound intensity was
determined according to the Margulis–Maltsev procedure [22].
Experiments were carried out in two modes: at maximum
ultrasound intensity (series 1) and at 80% of the maximum level
(series 2). Each series consisted of four experiments. As
cavitation fluids, a series of aromatic oligomers of formula
(n = 18–36; m = 14–26; x = 2–5) were synthesized.
This oligomeric series was chosen for its low saturated vapour
pressure and rather high boiling temperatures. Also, some
compounds from this series are used as the stationary phase in
gas chromatography, and therefore information about their
condensing coefficients at the fluid-gas boundary (α ∼ 1) is
known [23]. For the selected compound (an isomeric mixture of
five- and six-ring polyphenyl ethers, with the mass fraction of
the six-ring polyphenyl ethers equal to 11%), a suspension was
made with powdered (100–200μm) special purity graphite. The
total amount of inorganic impurities present in the graphite was
∼ 3 × 10
− 3
mass%, see Table 1. The solid–fluid weight ratio for
the suspension was 1:6 in all experiments, and the cavitation
fluid density was 1.1g cm
− 3
at 25°С. The reactor was filled
(52ml) with the suspension via the charging hole.
Prior to synthesis, the suspension was degassed by pumping the
system through the charging hole using a vacuum pump to a base
pressure of 20mTorr while simultaneously initiating the ultrasound
at 20% of maximum power for ∼ 10min. During this degassing
process, an intensive release of gas bubbles was observed, and the
reaction mixture increased in temperature to 80–100°С.
Diamond synthesis was started with the reaction mixture
temperature at 80°С by smoothly increasing the ultrasonic power
up to its maximum value. This power increase was accompanied
by the formation of cavitation zones over the whole reactor
volume, accompanied by a characteristic crackling sound. The
synthesis process was terminated when the reaction mixture
temperature reached 120°C, typically after 2min for the given
reactor cooling mode. Some experiments were carried out for
longer times (up to 12min). Note that the temperature quoted
above is that of the bulk reaction mixture; the local reaction
temperature will, of course, be much higher, of the order of many
thousands of K. Ideally, we would like to characterize the growth
conditions by measuring the local temperature, pressure and size
of the cavitation bubbles. These experiments are currently
underway and the results will be presented in a subsequent
After each experiment, the liquid phase was allowed to cool,
then removed from the reactor. It was then separated from the
solid synthesis product by centrifuging. The solid phase was
washed using acetone, then a control sample was collected, and
the remainder was oxidized using sulfochromicmixture (H
acids [24]) to remove any non-diamond carbon phase.
The diamond yield was determined by weighing the solid before
and after oxidation. Losses of the diamond phase during
separation did not exceed 5%. To accumulate sufficient synthesis
product for further analysis, samples from four identical
experiments in each series were mixed together.
The solid-state products of synthesis were examined using
X-ray analysis (URD-6 X-ray diffractometer K
wavelength 1.544Å), scanning electron microscopy, SEM
(JEOL 2010 for series 1, REM-100Y for series 2), and laser
Raman spectroscopy (Renishaw 2000 Raman Spectrometer).
The content of micro-impurities was determined by an emission
spectral analyzer DFS-13. The size distribution of the
synthesized diamonds was determined by software (ImageJ
v.1.38) analysis of SEM images of these diamonds.
3. Results and discussion
In all ultrasound experiments using the graphite powder
suspension, we observed formation of a solid product that was
not oxidisable by sulfochromic mixture. X-ray analysis of this
non-oxidisable product (Fig. 2) produced peaks corresponding
to d-spacings of d = 2.058Å and 1.262Å, 1.0752Å, which are
consistent with literature values for the cubic diamond unit cell
(e.g. 2.06Å, 1.26Å and 1.075Å for the {111}, {220} and {311}
planes, respectively [25]). Also shown in Fig. 2 is the X-ray
pattern from the initial graphite, for comparison.
During the 2min cavitation, the yield of diamond in all
experiments was ∼ 0.5% of the initial graphite mass for series 1,
and ∼ 2% for series 2. Fig. 3 shows SEM images of the
synthesized diamond particles. Photomicrography analysis
reveals that each particle is a shaped microcrystal with
distinguishable growth faces. The particle form factor was 1.1
to 1.2. Fig. 4 presents a distribution bar chart of the diamond
particle sizes for both series. For all experiments, nearly
monodispersed diamond powder was obtained, with series 1
producing 6μm crystallites and series 2 producing 9.5μm
crystallites, with a size spread in both cases of ∼ 0.5μm. This
spread may simply be a result of mixing the diamond products
from 4 experiments, where weak fluctuations of electrical
Table 1
Most abundant inorganic impurities (units of % by mass) present in the initial
graphite and in the diamond product, as measured by an emission spectral
analyzer DFS-13
Initial graphite
− 4
− 4
− 4
− 4
− 3
− 3
− 4
− 3
− 4
− 4
− 4
− 4
− 5
− 4
A.K. Khachatryan et al. / Diamond & Related Materials 17 (2008) 931936
parameters were inevitable. Alternatively, the spread may be a
result of inaccuracies in the software analysis.
Of particular interest is the correlation between particle sizes
in each series with the diamond phase yield. Mathematical
processing of the diamond phase yield data and the particle size
(in view of equal particle size in each series) in series 1 and 2
allows us to calculate that, in each experiment, an equal amount
of diamond particles is formed.
An apparent conclusion from these experimental results is
that the diamond formation process is of probabilistic nature,
i.e. that it is controlled by the number of “successful” (resulting
in the diamond phase formation) cavity collapses. We believe
these occur when the shockwave is directed normally to the
randomly-oriented graphite basal planes. Each particle of the
graphite (sized 100–200μm) is a sintered agglomerate consist-
ing of randomly oriented crystallites. We believe that each
“successful” collapse is a cavitation implosion directed nor-
mally to the basal planes of separate individual crystallites.
Since there are many crystallites oriented along different
directions, there is a high probability that any given cavitation
collapse will be normal to at least one of these crystallites,
resulting in high yield. To confirm this, an experiment was
performed under identical conditions but using quasi-mono-
crystalline graphite (particle size ∼ 100–200μm) having highly
oriented basal planes. For these, the probability of a cavitation
collapse being normal to the graphite particle surface is much
lower, since on average the crystals will be oriented in the
wrong direction far more often than in the normal direction. As
expected, the cavitation experiment now formed diamond only
in trace quantities, insufficient for analytical examination. This
suggests that the graphite-to-diamond transformation results
from the collapse of a single cavitation bubble that formed on
the graphite surface (one “successful” collapse = one diamond
The form of the particle size distribution curve is related to
their formation mechanism [26]. The very narrow spread of
particle sizes (± 0.5μm) observed in both growth series suggests
that the synthesis is terminated rapidly after the shock wave has
passed. This is because the formation of monodispersed
diamond crystallites sized up to 10μm in such short time
periods cannot be explained by the existing models of diamond
Fig. 3. SEM images of the cavitation diamonds: (a), (b) series 1 (taken at Bristol); (c), (d) series 2 (taken at Yerevan).
Fig. 2. X-ray patterns of (1) the initial graphite and (2) the solid, non-oxidisable
product after cavitation. The peaks in (2) at 2θ=44° and 2θ=76° correspond
closely to those expected for diamond peaks with d-spacings 2.06 Å (111 plane)
and 1.26 Å (220 plane), respectively [25].
A.K. Khachatryan et al. / Diamond & Related Materials 17 (2008) 931936
cluster growth [27]. At extreme conditions diamond formation
must occur by another mechanism.
Table 1 presents data on the impurity composition for seven
elements contained in both the initial graphite and the
synthesized diamonds. Analysis shows that the impurity
concentrations of the synthesized diamond are almost identical
to those of the initial graphite (although an increased content of
potassium and chromium in the diamond is related to the acid
treatment during product separation).
Fig. 5 presents Raman spectra of (a) the initial graphite, (b) the
unpurified graphite/diamond product following cavitation, and
(c) the separated diamonds, for series 2, using green (514 nm)
laser excitation. For the starting material, as expected, the Raman
spectrum shows only the 1580 cm
and (small) 1380 cm
peaks characteristic of ordered and disordered graphite, respec-
tively. For the unpurified cavitation product, both the 1580 cm
peak and the 1332 cm
peak (characteristic of crystalline
diamond) appear. For the purified diamond product only the
1332 cm
peak is present. This shows that the purification
process was successful in removing virtually all the sp
material from the product, leaving pure diamond. The rising
photoluminescent background is typical of small-grained dia-
mond and results from defects in the crystallites and scattering
from crystal edges. It is also worth mentioning that the Raman
spectra were easily capable of detecting the presence of the
diamond phase within the cavitation product at a concentration of
1.5%. This suggests that Raman analysis might prove an
invaluable tool for process optimization and control when
developing this synthesis method further.
As well as initiating diamond formation, the cavitation process
‘crushes’ the graphite into smaller particles, which appear to be
less efficient at making diamond. To investigate this, experiments
were performed using various sizes of graphite (0–50, 50–63,
100–150, 100–200, and 100–250 μm). Diamonds were only
obtained for the largest three sizes. After the synthesis the solid
phase was separated out, dried and sieved (100 μm hole size) to
determine what fraction of the graphite had been crushed into
smaller particles. After a synthesis time of 2 min, we found that
∼5% of the graphite (initial size 100–200 μm) had been crushed,
and passed through the 100 μm sieve. This 5% was oxidized
separately, but after analysis found to contain no diamonds.
Diamonds were found, however, after oxidizing the larger
graphitic fraction (that did not pass through the sieve). This
suggests that diamond formation requires large (N60 μm)
graphitic particles, and that the diamond may have formed inside
or at the surface of these particles. However, SEM analysis of the
graphite after the synthesis was inconclusive, and the question
remains open. One possible explanation for the size effect is that
the kinetic energy of the shockwave is not absorbed by small mass
graphitic particles, and ‘by-passes’ them, whereas the more
massive particles, with their higher inertia, experience the full
shock wave.
A consequence of this, however, is that continuation of the
cavitation process beyond a few minutes is unfortunately not
accompanied by an increase in the diamond yield. This is because
as cavitation proceeds, the graphite particle size reduces and the
efficiency of diamond production drops. However, manifold
increases in yield can be achieved by completely refilling the
cavitation medium while maintaining the solid phase. The
maximum yield of microdiamonds for the 10 μm fraction was
10 mass% after 6 complete refillings of the cavitation fluid.
Hence, the limiting stage of cavitation synthesis is the change of
the cavitation medium chemical composition. Note that the
cavitation medium is easily regenerable by pumping down, and
that it can be used repeatedly.
Unfortunately, it is not yet possible to specify an optimum
composition (initial or intermediate) of the cavitation media for
reaching the maximum cavitation impulse, and therefore for
diamond formation. It is also possible that easily volatile and
gaseous products from the destruction of the cavitation medium
may result in a premature elastic collapse of the cavities and
attenuate the cavity impulse.
It is worth comparing the energy efficiency of this cavitation
process with other methods of making diamond. Our ultrasonic
reactor delivers 1 kW for a total reaction time of ∼2 mins,
which translates into a total energy delivered to the reaction
chamber of 120 kJ. Since ∼10 g of graphite was initially present
in the reactor and (for the 9.5 μm fraction) 2% of this was
Fig. 5. Laser Raman spectrum (514 nm) of (a) the graphite starting material,
(b) the unpurified product containing ∼1.5% diamond in graphite, and (c) the
purified (100%) diamond product. The spectra have been offset vertically for
Fig. 4. Size distribution bar chart for diamond particles from both series. Both
series contained a mixture of the products of four identical experiments.
A.K. Khachatryan et al. / Diamond & Related Materials 17 (2008) 931936
converted to diamond, the total yield was ∼0.2 g of diamond
(1 carat). Therefore the cavitation process requires 600 kJ per
gram of diamond. Compare this to a typical 1 kW microwave
CVD reactor, which grows, say, 5 μm of diamond over an area
of 1 cm
in an hour. This would have an energy usage of
2000 MJ g
− 1
of diamond. Similarly, for HPHT diamond, the
efficiency is estimated to be ∼400 kJ g
− 1
of diamond. This
shows that ultrasonic cavitation is a very energy efficient route
to forming diamond compared to CVD, and comparable to that
for HPHT. If developed successfully, this technology, therefore,
has the potential to be a realistic commercial prospect.
4. Conclusions
We have shown that, although at an early stage of develop-
ment, the cavitation synthesis method opens new opportunities in
obtaining diamond crystallites at relatively benign conditions,
average bulk temperature of the liquid up to 120 °C and at
atmospheric pressure. It produces crystalline diamonds with a
very sharp, well defined size range of 5–10 μm, and with high
purity. The major factor influencing diamond yield in the
cavitation process is the composition of the cavitation medium,
and there is wide scope here for future experiments. Furthermore,
the rapid timescale for diamond formation may allow controllable
doping with selected impurities such as N or B, with a view to
obtain p- and n-type semiconducting diamonds. An optimum
method of cavitation generation (pulse or periodic signal) may
enable us to maintain a constant composition of cavitation
medium and, consequently, to increase the diamond yield
considerably. The absolute size of the synthesized diamond
powders and their size distribution are related to the synthesis
conditions (p and Т). Hence, it is possible to determine
quantitative energy characterization of cavitation perturbations.
Our present cavitation module allows up to 10% graphite-to-
diamond conversion in a continuous regime of regeneration for
the cavitation medium, thus opening commercial prospects for
the method. Future designs could increase this yield further.
With suitable choice of cavitation media, the cavitation method
could be used for synthesis of other potentially important
compounds, such as cubic boron nitride, carbon nitride, etc.
The authors are grateful to all who rendered technical, material
and moral support during the research. Specialgratitude should be
expressed to Dr V. Khachatryan, a specialist in organic synthesis,
for the help in the cavitation media synthesis, and to Prof. N.
Ananikyan, a specialist in theoretical physics, for useful advices

vB Professionals