E. Ashkinazi
580 California St., Suite 400
San Francisco, CA, 94104
Single crystal diamond growth by MPCVD at subatmospheric pressures
2020, Materials Today Communications
https://doi.org/10.1016/J.MTCOMM.2020.101635
Sign up for access to the world's latest research
Abstract
Microwave plasma assisted chemical vapor deposition (MPCVD) is the established technique to produce high quality single crystal diamond (SCD). While typical pressures for SCD growth regimes in methane-hydrogen plasma are currently within 100-300 Torr, a transition to much high pressures promises enhanced growth rates. Here, we report on successful SCD synthesis by MPCVD in CH 4-H 2 gas mixtures at pressures up to 600 Tорр. A strong change of the plasma shape and volume (the latter shrinks by 10 times at fixed MW power) with pressure rise from 100 to 600 Torr was observed, still keeping the plasma stable. The record high absorbed MW power density of ≈1800 W/cm 3 was achieved at 600 Torr. Optical emission spectroscopy (OES) was used for the plasma analysis via monitoring emission intensities of radicals H α , C 2 and CH. The gas temperature T g determined from analysis of rotational fine structure of OES Swan transitions of dimer C 2 (516 nm) turned out to be essentially constant ~ 3100 ± 150 K over the pressure range explored. The diamond growth rate is found to increase by an order of magnitude with pressure to achieve 57 μm/h at 500 Torr at relatively low (4%) CH 4 concentration, as measured in situ using low-coherence interferometry, but declined at further pressure increase. The produced films were characterized with SEM, XRD, Raman and photoluminescence spectroscopy, and a high/moderate quality of the obtained material was confirmed.
Key takeaways
AI
AI
- SCD growth rates reach 57 μm/h at 500 Torr, declining post 600 Torr due to plasma shrinkage.
- MPCVD achieved record MW power density of 1800 W/cm³ at 600 Torr, enhancing growth conditions.
- Gas temperature remains constant at ~3100 ± 150 K across pressures, indicating stable plasma conditions.
- Plasma volume reduces by an order of magnitude from 100 to 600 Torr, affecting growth efficiency.
- The study emphasizes establishing a common protocol for defining plasma dimensions to enhance comparability.
Materials Today Communications 25 (2020) 101635
3
passing only H
α
line emission (bottom raw) and without the fltering
(top raw). The plasma cloud progressively shrinks with pressure in-
crease, its shape transforming from a sphere at 100 Torr to a strongly
elongated vertically ellipsoid at 600 Torr. The plasma volume was
assessed from the "red" images (with H
α
flter) according to procedure
described in Ref. [20].
The spatial profle of H
α
line intensity I
Hα
(x,y,z) assesed from the
“red” image have been found to be very close to Gaussian profle, so the
plasma has the shape of ellipsoid of revolution:
I
Hα
= I
0
× exp[– (x
2
+y
2
)/2S
x
2
] × exp[–z
2
/2S
z
2
]
where X
e
= √2S
x
(Y
e
= √2S
x
) and Z
e
= √2S
z
are radii from the plasma
center in the three directions x,y,z, respectively, at which the intensity
reduces to 1/e level (e = 2.718). This gives the plasma volume
V = 4πX
e
2
×Z
e
/3, used to estimate the absorbed power density P/V. Un-
fortunately, still there is no a common protocol how to measure the
plasma size and volume. To defne the effective dimensions of the
plasma we used cutoff value I
cut
/I
0
= 1/e throughout the present paper.
Some authors choose the cutoff at the emission intensity level of 0.66
using a discharge photo without any flter [2], while others took a photo
with H
α
flter and adopted the cutoff level of 0.15 (≈1/e
2
= 0.135) [21].
Such variation of the emision cutoff level results in signifcant difference
in the estimated volume and MWPD values. As an example, relative
calculated diameter D and P/V~D
–3
for a spherical plasma with
Gaussian profle for emission intensity for different defnitions of the
plasma size (different choice of the I
cut
/I
0
level) are given in Table 1.
Here, the diameter D was normalized to the value D
e
defned for the case
of I
cut
/I
0
= 1/e, so D/D
e
= 1 as follows from relation I
cut
= I
0
× exp
[–(D
e
/2)
2
/2S
2
], and D/D
e
= [–ln(I
cut
/I
0
)]
–1/2
for a general choice of
I
cut
/I
0
ratio. One can see that the MWPD value is overestimated by 3.7
times when the cutoff I
cut
/I
0
is 0.66 [2], and is underestimated by a factor
of 2.6 for cutoff I
cut
/I
0
= 0.15 [21]. This underlines again the importance
to establish a common procedure for the plasma size evaluation.
The diameter 2X
e
and vertical length 2Z
e
defned from intensity of H
α
line emission profle with cutoff level 1/e of the maximum intensity I
0
, as
described above, both decrease with pressure but with different rates as
shownin Fig. 3. The diameter reduces by a factor of three from 35 mm
(p = 100 Torr) to 12 mm (600 Torr), while the vertical size decreases to a
smaller extent, by a factor of two only, from 25 mm to 13 mm. Using the
same approach for the plasma dimensions assessment, but based on the
images taken with the green flter to map the emission of C
2
line at
516 nm, very similar plasma size evolution was observed (Fig. 3). The
difference in sizes of “red” and “green” plasma images are less than
2 mm both for 2X
e
and 2Z
e
values, the size of the “red” image always
being larger. If the plasma size is defned using the green emission of
dimer C
2
, this would underestimate the plasma size roughly by 10 % in
the worst case, and the plasma volume by ~30 %. Note, that for low
pressure of 100 Torr the difference in the “red” and “green” image sizes
almost vanishes.
Interestingly, that while shape and size of the plasma strongly
changed with pressure, its position relative to the substrate holder was
quite stable. The positions of the plasma bottom and top (again, the
measuring points are defned at H
α
line intensity cutoff level 1/e) and
location of the center with maximum plasma emission, are essentially
fxed with respect to the holder top surface with the pressure variation as
displayed in Fig. 4. The variations in position of the plasma bottom were
within 1 mm at most across the entire pressure range from 100 to
Fig. 2. Photographs of the discharge in 4.1 %CH
4
– H
2
gas mixture above 4 × 4 mm
2
SCD substrate inside a pocket substrate holder at pressure increasing from 100 to
600 Torr. The images are taken without (top row) and with red flter (bottom row) passing only H
α
line emission (λ = 656.5 ± 6 nm) in the plasma spectra. At the
bottom of some photos, plasma refection in the holder top surface can be seen. Microwave power P = 2.7 kW and total fow rate of 417 sccm were kept constant.
Table 1
Relative diameter D and absorbed microwave power density P/V~D
–3
for a
spherical plasma with Gaussian radial profle of optical emission intensity
calculated for fxed power at different cutoff levels I
cut
/I
0
of plasma brightness
along X-axis. The values D and P/V are taken as 1.0 at cutoff I
cut
/I
0
= 1/e.
I
cut
/I
0
* D P/V~D
–3
Ref.
0.66 0.65 3.7 [2]
1/e≈0.368 1.0 1.0 this work
0.15 1.38 0.38 [21]
Fig. 3. The diameter 2X
e
(full symbols) and vertical length 2Z
e
(open symbols)
of the plasma at different pressures, as determined using the plasma images
taken with red flter (H
α
line emission, 652 nm, circles), and green flter (C
2
line
emission, 516 nm, squares). The plasma borders are defned by locations where
the intensity drops to level of 1/e relative to maximum intensity I
0
. Microwave
power P = 2.7 kW, 4.1 % CH
4
.
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
4
600 Torr. We note that, since the substrate was moved in Z-direction for
a distance in a narrow range within Δd<0.6 mm to stabilize the sub-
strate temperature (see Section 3.2 and Fig. 10), the Fig. 4 also indicates
approximately constant position of the plasma bottom with respect to
the substrate. In contrast, the top edge of the plasma cloud (and,
consequently, the vertical size 2Z
e
) lowered by 6 mm with pressure. The
plasma positions almost identical to those in Fig. 4 have been deduced
from optical images taken through green flter (not shown here).
Fig. 5 shows vertical profles of plasma optical emission intensities
I
Hα
(Z) and I
C2
(Z) of H
α
and C
2
spectral lines, respectively, across the
substrate center (X = 0) at pressures of 100, 300, 500 Torr deduced from
photographic images taken through “red” and “green” flters. The pro-
fles I
Hα
(Z) and I
C2
(Z) reveal similar width at a given pressure. Also,
asystematic displacement of the H
α
line profle towards the substrate
compared to the C
2
counterpart profle takes place, better seen for lower
pressures of 100 and 300 Torr (Fig. 5a,b). Similarly, such shift along Z
axis between I
Hα
(Z) and I
C2
(Z) maxima was observed earlier using a
spatially resolved OES by Ma et al. [22] at 150 Torr, Bogdanov et al. [24]
at 145 Torr, and Bolshakov et al. [6] at 130 Torr. The shift, however, was
different from the present data due to a difference in process parameters,
such as gas mixture composition, cavity geometry and the electromag-
netic feld modes involved in CVD systems used. At higher pressure
(500 Torr) a part of the C
2
emission profle, from substrate to the peak,
fully coincides with that for H
α
profle, but has a shorter tail towards
larger distances Z from the substrate (Fig. 5c).
With the data on the plasma size (Fig. 3) obtained from the images
taken with red flter (656 nm) the plasma volume of ellipsoid shape and
average microwave power density, were calculated as functions of
pressure (Fig. 6). The plasma volume V decreases by an order of
magnitude, from 17 to 2 cm
3
when the pressure increases from 100 to
600 Torr, while the absorbed microwave power density P/V, increases
from 180 W/cm
3
up to 1800 W/cm
3
. To our knowledge the highest so far
reported MWPD values for CVD diamond growth did not exceed
1000 W/cm
3
for 2.45 GHz plasma [5] and 1500 W/cm
3
for 28 GHz
plasma [25]. We defned absorbed microwave power P
abs
as incident
power P
ins
minus refected power P
ref
, the latter two values are measured
directly. The refected power was low (0.4 – 0.5 % of P
ins
), slightly
increasing to 30 W (≈1% of incident power) upon transition from 500 to
600 Torr. The 3D modeling of electromagnetic absorption using the CST
Microwave Studio software shown the ohmic losses in waveguide,
mode-transformer and reactor cavity to be of 2–3 % of the incident
power (electric properties of components made of aluminum and
stainless steel were taken into account), so these losses were neglected as
well. Therefore, the contribution of the error in the power to overall
error in MWPD is much smaller than that of the plasma volume
inaccuracy.
Fig. 4. Positions of bottom edge (triangles), center (circles) and top edge
(squares) of the plasma cloud with respect to the holder top surfaceat different
pressures, as determined from “red” images with 1/e cutoff level. The
“maximum” is the location of the maximum plasma emission intensity. Mi-
crowave power P = 2.7 kW, 4.1 % CH
4.
Fig. 5. Vertical profles of plasma emission intensities I
Hα
(Z) (solid line) and I
C2
(Z) (dashed line) at pressures of 100, 300 and 500 Torr obtained from images taken
through red and green flters, respectively. Z = 0 corresponds to the holder top surface. The profles are normalized to maxima values.
Fig. 6. Plasma volume V (squares) and absorbed microwave power density
MWPD (circles) vs pressure in 4.1 %CH
4
-H
2
mixture. Microwave power
is 2.7 kW.
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
5
3.2. Optical emission spectroscopy
Panoramic optical emission spectra taken in the wavelength range of
400–750 nm at different pressures (Fig. 7) demonstrate a group of lines
typical for MW plasma in CH
4
-H
2
mixtures [22,26], including atomic
hydrogen H
α
(656 nm), carbon dimer C
2
(transitions with change in
vibration levels Δν = –2; –1; 0; +1; +2) and CH radical (at 314 and
431 nm), the C
2
(Δν = 0) line dominating the spectra. Note, that the
diameter of the region in the plasma from which the OES signal was
recorded, did not exceed 1.5 mm and was close to the center of the
plasma, i.e. always, even at the highest pressure of 600 Torr, it was less
than the size of a plasma cloud. The intensities of all the emission lines
show a strong increase with pressure, refecting the observed increase of
the plasma brightness.
The plot for intensity of H
α
, C
2
and CH lines in the spectra as a
function of pressure is displayed in Fig. 8. The brightest line C
2
(516 nm)
enhanced in intensity by approximately 70 times with pressure increase
from 100 to 600 Torr. It is this emission line (the leading sharp peak
accompanied by a broad rotational fne structure) that gives rise to reach
green color of the plasma emission (Fig. 2). The intensities of two CH
lines at 413 and 431 nm increased by factor of 20 and 80, respectively,
while only a moderate, ≈4 times, increase in H
α
line intensity was
observed. Note that the OES signal for different species was taken from
the location where the H
α
intensity showed a maximum. Because of a
relative shift of C
2
(Δα = 0) and H
α
intensity profles along Z direction
(actually, the shift was seen at low pressure of 100 Torr only, see Fig. 5a)
the datapoints of intensities for C
2
line in Fig. 8 on are somewhat lower
than those in the C
2
maximum at given pressure. However, one can see
in Fig. 5a that the C
2
and H
α
intensity profles have rather fat tops,
therefore the C
2
intensity is underestimated by a few percents (ca. 3%)
only. For this reason this defcit was neglected.
The enhancement of the emission of atomic hydrogen and carbon-
related radicals indicates an increase in concentration of those species
with pressure (and MPWD) that should lead to higher growth rates [27,
28]. However, at pressures above 500 Torr the emission intensity either
decline as for CH and H
α
, or saturate, as for C
2
. This observation cor-
relates with a similar decline in growth rate at 600 Torr as will be shown
below.
The gas temperature was determined as rotational temperature by
analyzing the fne structure of Swan transitions for dimer C
2
emission
(516 nm, Δν = 0) according to procedure described elsewhere [20,23].
The dependence of the gas temperature T
g
on pressure is shown in Fig. 9.
It seen that the gas temperature is almost constant T
g
~ 3100 ± 150 K
over the pressure range explored, with a slight decline above 400 Torr.
We speculate that increasing energy loss of the plasma at higher pres-
sures could be due to enhanced thermal conductivity loss for compressed
plasma ball as its surface-to-volume ratio increases. As a side effect, a
small increase in refection power of 30 W (≈1% of incident power) was
noticed upon transition from 500 to 600 Torr. The gas temperature in
methane-hydrogen plasma at pressures above 350 Torr was measured
here for the frst time.
We note that Derkaoui et al. [28] reported a monotoneous increase of
gas temperature for MW discharge in CH
4
-H
2
mixture from 2150 K to
3000 K with pressure rise from 20 to 200 Torr, as determined both with
OES and coherent anti-Stokes Raman scattering (CARS). However, they
maintained the plasma volume constant by increasing the MW power by
a factor of seven to prevent the plasma compressing, while, in contrast,
we kept the power constant. It is important that in our case, despite of a
strong increase of absorbed MW power density in the plasma with
pressure p (see Fig.6), the absorbed power per particle in the plasma
P/Vn, where n is total concentration of neutral and charged particles, is
proportional to P/Vp (as a rough model we consider a uniform plasma
cloud), and remains essentially constant for pressures of 200–500 Torr
as shown in Fig. 9. Only on the terminal pressures of 100 and 600 Torr,
the MPWD/p value shows some increase. A deviation of low-p datapoint
from horizontal dependence has no clear explanation at the moment
(may require plasma simulation). The MPWD/p increases at 600 Torr for
Fig. 7. Panoramic OES spectra for discharge in 4.1 %CH
4
+ H
2
at pressures
200 Torr (red spectrum), 400 Torr (yellow spectrum) and 600 Torr (green
spectrum). The spectrum for 200 Torr is multiplied by 5 times.
Fig. 8. The intensities of emission lines C
2
(Δν = 0), H
α
and CH (at 314 and
431 nm) in OES spectra depending on pressure. The intensities are not corrected
to the spectrometer spectral sensitivity. Process parameters: 2.7 kW, 4.1 % CH
4.
Fig. 9. Dependence of gas temperature T
g
(squares) and MWPD/p (circles) on
pressure p. Process parameters: 2.7 kW, 4.1 % CH
4.
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
6
40 %, however without a corresponding increase in gas temperature T
g
.
A possible reason for the lag between MWPD/p and T
g
at 600 Torr could
be an enhancement of energy loss from plasma via thermal conductivity
and radiation loss upon further reduction of the plasma size, as the
plasma surface area-to-volume ratio rises. Here n stands for total number
of species, neutral and charged radicals and molecules, in the unit
plasma volume (as a rough model we consider a uniform plasma cloud).
As the particles concentration increases with pressure, the total number
of the particles in the plasma does not change since the plasma is
self-consistently compressed in such a way that the absorbed MW power
per particle remains almost unchanged with pressure variation. There-
fore, the gas temperature T
g
can be expected to be only slightly depen-
dent of pressure, this being in agreement with our observation.
3.3. Growth rate
The growth rate was measured in situ using low-coherence interfer-
ometry during 1520 min, then the pressure was changed without the
plasma switch-off to repeat the measurement. The distance d between
the top surface of the substrate inside the pocket holder and top surface
of the latter (Fig.1b) was tuned upon the pressure change to keep the
sample temperature of 1050 ± 12
◦
C constant over the entire pressure
range explored. The pressure increased, frst, from 200 to 600 Torr, then
decreased back to check if any hysteresis in growth rate exists. The
measured dependence of growth rate vs pressure is shown in Fig. 10 a.
The growth rate rapidly increases with pressure, from 6 μm/h at
200 Torr to 57 μm/h at 500 Torr, but declines to ≈50 μm/h at further
pressure increase up to 600 Torr. Also shown on the plot are the data
points obtained upon the pressure decrease in steps from 600 to
200 Torr, which generally reproduce the values measured at the pres-
sure increase.The hysteresis in the growth rate is relatively small except
only the terminal datapoint for 200 Torr. We suggest that a possible
reason for the difference in “forward” and “reverse” growth rates could
be a difference in surface relief for the 200 Torr sample at the beginning
of measurement (we started from deposition at 200 Torr of thin layer on
smooth polished HPHT substrate) and at the end of process (a typical
step-and-terrace surface structure was well developed when up-down
pressures cycle was fnished). The effect of the surface relief
morphology on growth rate was reported recently by Shu et al. [29].
Typically a rise in T
g
is provided by increase of MWPD [28]. The dia-
mond growth rate is expected to increase with gas temperature T
g
due to
enhanced generation of atomic hydrogen and CH
n
radicals, such as
methyl radical CH
3
[30]. In our case the growth rate increases with
pressure at constant T
g
simply due to increase of concentration of radicals
with gas density increase.
The dependence of distance d between the substrate surface and
holder top surface on pressure is given Fig. 10b. We note that it was
suffce to vary the depth d in a narrow range of 3.7–4.3 mm to be able to
stabilize the substrate temperature within the entire pressure range
explored. The substrate was progressively moved deeper in the pocket
with pressure rising up to 500 Torr, but at the highest pressure it was
moved up by 0.1 mm to d = 4.2 mm. This indicates that the thermal fux
from the plasma somewhat reduced at 600 Torr. The plot in Fig. 10a is
actually a result of combined effects of pressure and sample depth d,
which generally is not easy to split. We note that Charris et al. [31]
measured the growth rate as function of substrate depth in a pocket-type
substrate holder keeping the substrate temperature and MW power close
to the parameters in our work, but at a moderate pressure (240 Torr).
They found a change in the growth rate of about 3% (see Table 2 in [31])
upon change of Δd = 0.3 mm in the substrate depth d, that translates to
only ~1% for Δd = 0.1 mm. In the present experiment, while increasing
the pressure from 500 to 600 Torr the growth rate drops from 57 μm/h to
50 μm/h, that is by 14 %, an order of magnitude larger than could be
expected for Δd = 0.1 mm if the data from [31] could be applicable for
our CVD reactor. However, in view of difference in designs of the re-
actors and holders, the substrate positions in the holder and other de-
tails, we can not exclude a larger effect of the depth d on the growth rate
in our case compared to results by Charris et al. [31].
A similar effect of growth rate fattening with pressure, but at lower
value of 400 Torr, was observed previously by Muehle et al. [2]
(Fig. 10a). They ascribed this limitation of the growth rate to an increase
of the distance between the discharge and the substrate as the pressure
increases, in other words, the discharge was pulled away the substrate.
Our observation, however, does not confrm such moving of the plasma
(see Fig. 4). A possible reason for the saturation of the growth rate fol-
lowed by decrease with pressure could be a reduced supply of radicals
from the plasma sides (in radial direction) in proximity to the substrate.
While radial density above the substrate continues to increase with
pressure, the plasma diameter becomes comparable with the substrate
size, so the lateral fux of species reduces. Therefore, in total budget of
species consumed for growth the contribution of radicals arrived from
the plasma sides decreases when the plasma shrinks too much.
One more potential factor reducing the growth rate could be a
decrease in effective nitrogen impurity concentration in gas due to a
leakage in the reactor. The leakage often is a main source of nitrogen
Fig. 10. Growth rate vs pressure in 4.1 % CH
4
+H
2
gas mixture as measured in
situ upon pressure increasing from 200 Torr to 600 Torr (open circles) and
subsequent decreasing from 600 Torr to 200 Torr (closed circles) without
plasma switch off at MW power of 2.7 kW. The solid curve is a guide for eye.
The substrate temperature 1050 ± 12
◦
C was stabilized by changing the depth
d of the substrate top surface with respect to the pocket holder top (see Fig. 1
for geometry). Squares are experimental data from Muehle et al. [2] (a).
Dependence of distance d needed to keep the constant substrate temperature vs
pressure. The solid curve is a guide for eye (b).
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
7
contamination in the system provided the purity of process gases is high
enough. It is known [32–35] that the presence of nitrogen in CH
4
-H
2
mixtures can strongly increase the growth rate of (100) face. The ni-
trogen fux though a leakage is proportional to difference p
0
– p between
the atmospheric pressure p
0
and reactor pressure p. The relative N
2
fux
from the leakage (p
0
– p)/p
0
decreases by a factor of four from 0.87 at
100 Torr to 0.21 at 600 Torr (and goes to zero at p
0
= 760 Torr),
resulting the growth slow down while N
2
is removed from the reactor.
Therefore, the observed fattening for growth rate with pressure and
even a tendency to slow down with further rise of the pressure could be a
combined effect of the plasma shrinking and elevation, with possible
effect of nitrogen impurity decrease. As a complimentary consequence,
the reactors operating at atmospheric pressure would be able to
completely avoid the gas contamination due to leakage, and hence,
growth of extremely pure diamond could be easily realized.
3.4. Sample characterization
The diamond flms grown at different pressures are shown in Fig. 11.
The CVD diamond of 1.5 mm thickness was produced at 170 Torr at 3.5
% CH
4
for 107 h growth run without stops. In other experiments we were
able to maintain the growth process at similar conditions as long as for
330 h. The samples prepared at 500 and 600 Torr (Fig. 11b,c) with
growth rate of 54 and 51 μm/h, respectively, revealed smooth surface
without macro defects such as hillocks or polycrystalline inclusions.
Growth steps typical for epitaxial SCD were observed in optical micro-
scope (Fig. 11d).
X-ray diffraction confrmed monocrystalline structure of the pro-
duced samples. The XRD spectrum for the epitaxial diamond flm grown
at pressure of 600 Torr is shown in Fig. 12. The only observed refex at
angle 2θ = 119.495
◦
belongs to (400) plane. This refex consists of two
components for Kα1 and Kα2 lines with intensity ratio of 2:1. The
absence of refexes from other planes indicates single crystal structure of
the sample. The refexes from planes different from (400), if any, have
intensities at least three orders of magnitude lower compared to the
(400) refex as better seen on the spectrum in logarithmic scale (see
Supplementary Information, Fig. S1). The XRD spectrum for the sample
produced at moderate pressure of 170 Torr revealed very similar picture
with only a single peak of (400) plane around angle 2θ = 119
◦
indicating
the monocrystalline plate with (100) orientation (Supplementary In-
formation, Fig. S2).
The quality of the deposited diamond was assessed using Raman and
photoluminescence (PL) spectroscopy. To compare crystalline quality of
different samples the PL spectra for 1.6 mm thick sample produced at
500 Torr, 1.5 mm sample produced at 170 Torr (the growth rate for this
sample was relatively low which normally promises a better crystal
quality) and reference natural diamond are displayed in Fig. 12. The
spectra for 170 and 500 Torr samples reveal a narrow (2.7 cm
1
) Raman
peak and nitrogen-vacancy NV
◦
and NV
PL bands with zero-phonon
lines at 575 and 638 nm, respectively. Although these two CVD dia-
mond samples show the PL intensity of nitrogen-vacancy NV centers
stronger that the reference natural IIa type crystal, the PL intensity is
almost an order of magnitude lower than Raman peak intensity. In terms
of NV abundance the samples produced at moderate and high pressures
turn out quite similar. Vanishingly small peak of silicon-vacancy SiV
with ZPL at 738 nm could be also seen in the PL spectra of CVD samples,
the material being highly pure of Si contamination due to two reasons.
First, the cylindrical quartz window separating the MW waveguide and
reactor cavity is far from the plasma (it’s positioned below a stage and
substrate holder), this being the intrinsic advantage of the reactor con-
struction. Second, diamond deposition on Si substrates was never per-
formed in this reactor, thus preserving its cleanness. Also observed are
components of phonon band of defect with ZPL at 467.5 nm (2.651 eV)
tentatively attributed by Zaitsev [36] to a nitrogen-related defect con-
taining interstitial atoms. The PL with 467.5 nm was reported recently
by Hei et al. [37] for CVD diamond flms deposited with a DC-arc jet
system (Fig. 13).
Raman spectra taken in the frequency range of 500 – 2000 cm
1
for
several samples revealed the strong narrow peak at 1332 cm
1
charac-
teristic for diamond as shown in Fig. 14. On zoomed spectra, additional
three weak wide bands at ~400, 900 and 1500 cm
1
were observed for
all the samples studied (Fig. S4, Supplementary Information), which are
components of phonon band of luminescence of the defect with ZPL at
467.5 nm as evidenced from comparison with the respective PL spec-
trum (Fig. S5) [36].
To get a certain statistics local Raman spectra were taken at three
different locations within a central zone of about 2 mm in diameter on
each sample using 473 nm excitation wavelength. The averaged Raman
peak width (full width at half maximum - FWHM) was in the range of 2.6
– 2.8 cm
1
over the entire pressure range, with a slight tendency of the
1332 cm
1
peak broadening towards 600 Torr as displayed in Fig. 15a.
The spread in peak width over the sample area is relatively small,
Fig. 11. Examples of as-grown diamonds produced at different pressures: (a) 170 Torr, flm thickness of 1.5 mm; (b) 500 Torr, 1.6 mm; and (c) 600 Torr, 0.1 mm. The
flms are not separated from the substrates. Typical surface morphology of the crystals taken with optical microscope is shown for 500 Torr sample on the right (d).
Fig. 12. X-ray diffraction spectrum for epitaxial diamond flm grown at pres-
sure of 600 Torr. The refex at 2θ = 119,495
◦
belongs to (400) plane.
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
8
indicating rather good uniformity of the structure. These width varia-
tions could be caused by fuctuation of intrinsic stress within the crystal.
For example, the variation in the Raman peak position was in the range
of 1332.6-1332.8 cm
1
when measured in the three sites on 600 Torr
sample.
Note, that the limited spectral resolution of the spectrometric
apparatus (due to broadened solid state laser generation wavelength)
overestimated the measured peak width by a value of about 0.6 cm
1
compared to a true value. For example, the Raman peak width for highly
perfect IIa type natural diamond crystal (a reference sample) measured
by us with the same instrument gave FWHM = 2.3 cm
1
[38], while the
width was as low as 1.7 cm
1
, when measured using 488 nm wavelength
of Ar + ion laser with more narrow generation line. This result indicates
high/moderate quality of the crystal grown at the very high pressures
used in the present work.
To see in more detail how uniform is the diamond quality, and how
the diamond Raman peak width (FWHM – full width at half maximum)
varies across a macroscopic area, the Raman spectra mapping was per-
formed over the area of 0.5 × 0.5 mm
2
arbitrary chosen within central
part of the samples. The mapping procedure included recording the local
Raman spectrum in a certain spot, with further shift of the laser beam to
the another point with the step of 10 μm. Totally 2500 datapoints were
collected for 50 × 50 pixel array, the width of Raman peak at 1332 cm
1
was calculated for each spectrum, and a map of FWHM in pseudocolors
was built.
The spatial distribution of the Raman peak width for 600 Torr sample
is shown in Fig. 15b. The FWHM shows variation in the range of 2.70 –
2.95 cm
1
with average width value of 2.83 cm
1
. The spotty pattern
may indicate variations of local intrinsic stress. The zones with FWHM
close to the average value occupy more than a half of the map, they are
shown in green color. Interestingly, that the spread and the average
FWHM value are close to those found from the spectra locally taken in
three different sites (Fig. 15a). Thus the simplifed approach with only
several measurements in arbitrarily chosen locations turned out to be
reasonable and informative in this particular case. Similar FWHM maps
have been obtained for the samples produced at different pressures, they
are displayed in Figs. S4 and S5 (see Supplementary Information). The
average FWHM values were 2.68 and 2.54 cm
1
for the flms grown at
170 and 500 Torr, respectively, with a slight tendency to decrease with
the pressure decrease. Thus, combined analysis with XRD and Raman
spectroscopy identifed the produced layers and crystals as SCD of high/
moderate quality.
4. Concluding remarks
We demonstrated stable operation of MPCVD system with CH
4
-H
2
gas mixtures at pressures as high as 600 Torr. The non-stop growth for at
least 30 h at 500 Torr was performed, while at moderate pressure of
170 Torr continuous growth runs for up to 330 h haven been routinely
realized. A strong change of the plasma shape from a spherical at
100 Torr to vertically elongated ellipsoid at 600 Torr was observed at
fxed MW power, while the plasma volume shrank by 10 times. As a
consequence the record high MW power density of 1800 W/cm
3
was
achieved at 600 Torr. We pay attention to importance to establish a
common defnition of the plasma cloud dimensions in order to be able
correctly compare the plasma size and absorbed microwave power
density in experiments of different researchers. Particularly, the value of
plasma volume may vary by a factor of up to three depending on choice
of how location of the plasma borders is defned.
The optical emission spectra for 2.45 GHz microwave plasma in CH4-
H2 mixtures were studied for pressures above 350 Torr for the frst time,
and can be used in dense plasma modeling. The gas temperature T
g
determined from analysis of rotational fne structure of Swan transitions
of dimer C
2
optical emission (516 nm) turned out to be essentially
constant ~ 3100 ± 150 K over the pressure range explored, that was
explained by our fnding that the microwave energy input per particle in
the plasma remains almost unchanged with pressure. The growth rate of
single crystal diamond increased by an order of magnitude with pressure
to achieve 57 μm/h at 500 Torr, but declined to ≈50 μm/h at further
pressure increase to 600 Torr. Interestingly that the observed evolution
of intensities of H
α
, C
2
and CH lines in OES spectra with pressure, a
strong increase followed by a saturation, reminds the similar trend for
growth rate. The epitaxial flms produced at different pressures revealed
a similar high quality in terms of Raman diamond peak width with only
a slight tendency of the peak broadening at 600 Torr.
The obvious advantage of the realized extreme high pressure growth
regimes is enhanced diamond growth rate, however, there are also
drawbacks due to strongly compressed plasma. The geometry of shrunk
plasma cloud, extending vertically, is not optimal for utilization of
radicals generated in the plasma for reactions on diamond surface.
Because of the small plasma size, multiple (many substrates in one run)
growth of SCD is diffcult. Concerning more fundamental issues of the
high-pressure process, we want to mention the effect of extremely high
brightness of the compressed plasma. The plasma emission, especially
UV part of its spectrum, might infuence (i) the gas chemistry, (ii)
Fig. 13. PL spectra for 1.6 mm thick sample grown at 500 Torr (2.7 kW, 2.5 %
CH
4
) (blue spectrum), 1.5 mm thick sample grown at 170 Torr (4.4 kW, 2.9 %
CH
4
) (red spectrum), and reference IIa natural diamond (black spectrum).
Fig. 14. Representative local Raman spectra of the samples produced at 170,
500 and 600 Torr. The spectra are normalized to intensity of 1332 cm
1
peak.
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
9
diamond surface chemistry, thus changing the diamond growth reaction
kinetics. As a matter of fact, the Swan bands for C
2
dimer, corresponding
to transitions with Δn =+1 and Δn =+2 have photon energies of 2.6 eV
(470 nm) and 2.8 eV (440 nm), respectively, which exceed, for example,
the activation energy for etching of diamond by atomic hydrogen
Ea = 32–42 kcal/mole (Ref. [39]). The role of the high intensity plasma
optical emission in diamond growth kinetics could be an interesting
topic to study.
Our result is a step on the way to development of MPCVD reactors
operated at atmospheric pressure to enhance the diamond growth rate
and facilitate, to a certain extent, the technical maintenance. One more
advantage of such “atmospheric pressure reactor” (APR) would be
elimination of the problem of the reactor (cavity) leakage that is often
the main source of nitrogen impurity in gas and, consequently, in pro-
duced diamond. Therefore, synthesis of highly pure diamonds, such as
device grade single crystal, could be easier realized even using con-
ventional (without high vacuum system) APR machine adapted for
operation pressures slightly above 1 atm. We note, that CVD reactors
operated at 600 Torr as in the present work, become APR class systems
being installed in places located at a high enough altitude, in mountains,
as atmospheric pressure reduces with altitude above sea level. The
pressure of 600 Torr is the normal atmospheric pressure at altitude of
2000 m. Particularly, in the list of big highland cities are Mexico City,
Mexico (2150 m) and Santa Fe, New Mexico, US (2134 m) appropriate
for APR accommodation already now. With further increase of operation
pressure above 600 Torr the critical altitude required decreases
extending the number of potential settlements for APR technologies.
Author statement
Andrey Bolshakov: Design of experiment, Methodology, Editing.
Victor Ralchenko: Idea, Original draft preparation, Writing-
Reviewing. Guoyang Shu: Samples investigation, Editing.
Bing Dai: Investigation.
Vladimir Yurov: OES measurement.
Egor Bushuev: Diamond growth experiment.
Andrey Khomich: Raman spectroscopy.
Alexander Altakhov: Diamond growth.
Evgeny Ashkinazi: Substrates preparation.
Irina Antonova: OES spectra analysis.
Alexander Vlasov: Samples treatment.
Valery Voronov: XRD measurements.
Yuri Sizov: CVD reactor preparation, Software.
Sergey Vartapetov: Validation.
Vitaly Konov: Supervision.
Declaration of Competing Interest
The authors declare no confict of interest.
Acknowledgements
The work was supported by the Russian Foundation for Basic
Research (Grant No. 19-52-53019), National Science Fund for Distin-
guished Young Scholars (Grant No. 51625201), National Natural Sci-
ence Foundation of China (Grant No. 51911530123), and 1000 Talents
Program. G.S. thanks the support from the program of China Scholar-
ships Council.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mtcomm.2020.10
1635.
References
[1] F. Silva, J. Achard, O. Brinza, X. Bonnin, K. Hassouni, A. Anthonis, K. De Corte,
J. Barjon, High quality, large surface area, homoepitaxial MPACVD diamond
growth, Diam. Relat. Mater. 18 (2009) 683–697, https://doi.org/10.1016/j.
diamond.2009.01.038.
[2] M. Muehle, J. Asmussen, M.F. Becker, T. Schuelke, Extending microwave plasma
assisted CVD SCD growth to pressures of 400 Torr, Diam. Relat. Mater. 79 (2017)
150–163, https://doi.org/10.1016/j.diamond.2017.09.013.
[3] Y. Mokuno, A. Chayahara, Y. Soda, H. Yamada, Y. Horino, N. Fujimori, High rate
homoepitaxial growth of diamond by microwave plasma CVD with nitrogen
addition, Diam. Relat. Mater. 15 (2006) 455–459, https://doi.org/10.1016/j.
diamond.2005.11.046.
[4] Q. Liang, C.-s. Yan, Y. Meng, J. Lai, S. Krasnicki, H.-k. Mao, R.J. Hemley, Recent
advances in high-growth rate single-crystal CVD diamond, Diam. Relat. Mater. 18
(2009) 698–703, https://doi.org/10.1016/j.diamond.2008.12.002.
[5] Y.J. Gu, J. Lu, T. Grotjohn, T. Schuelke, J. Asmussen, Microwave plasma reactor
design for high pressure and high power density diamond synthesis, Diam. Relat.
Mater. 24 (2012) 210–214, https://doi.org/10.1016/j.diamond.2012.01.026.
[6] A.P. Bolshakov, V.G. Ralchenko, V.Y. Yurov, A.F. Popovich, I.A. Antonova, A.
A. Khomich, E.E. Ashkinazi, S.G. Ryzhkov, A.V. Vlasov, A.V. Khomich, High-rate
growth of single crystal diamond in microwave plasma in CH4/H2 and CH4/H2/Ar
gas mixtures in presence of intensive soot formation, Diam. Relat. Mater. 62 (2016)
49–57, https://doi.org/10.1016/j.diamond.2015.12.001.
[7] C.J. Widmann, W. Müller-Sebert, N. Lang, C.E. Nebel, Homoepitaxial growth of
single crystalline CVD-diamond, Diam. Relat. Mater. 64 (2016) 1–7, https://doi.
org/10.1016/j.diamond.2015.12.016.
[8] A.L. Vikharev, M.A. Lobaev, A.M. Gorbachev, D.B. Radishev, V.A. Isaev, S.
A. Bogdanov, Investigation of homoepitaxial growth by microwave plasma CVD
providing high growth rate and high quality of diamond simultaneously, Mater.
Today Commun. 22 (2020), https://doi.org/10.1016/j.mtcomm.2019.100816.
[9] Q. Liang, C.Y. Chin, J. Lai, C.-s. Yan, Y. Meng, H.-k. Mao, R.J. Hemley, Enhanced
growth of high quality single crystal diamond by microwave plasma assisted
Fig. 15. (a) Diamond Raman peak width (FWHM) for SCD produced at different pressures. The spectra were taken at 473 nm excitation wavelength at three different
locations on each sample. The solid line is a guide for an eye. Inset: Raman spectrum for the sample grown at 500 Torr. (b) The map of diamond Raman peak FWHM
over 0.5 × 0.5 mm
2
area for the sample grown at 600 Torr demonstrating variation in the width value in the range of 2.70 – 2.95 cm
1
.
A.P. Bolshakov et al.
Materials Today Communications 25 (2020) 101635
10
chemical vapor deposition at high gas pressures, Appl. Phys. Lett. 94 (2009),
https://doi.org/10.1063/1.3072352.
[10] A.P. Bolshakov, V.G. Ralchenko, V.Y. Yurov, G. Shu, E.V. Bushuev, A.A. Khomich,
E.E. Ashkinazi, D.N. Sovyk, I.A. Antonova, S.S. Savin, V.V. Voronov, M.
Y. Shevchenko, B. Dai, J. Zhu, Enhanced deposition rate of polycrystalline CVD
diamond at high microwave power densities, Diam. Relat. Mater. 97 (2019),
https://doi.org/10.1016/j.diamond.2019.107466.
[11] S.A. Bogdanov, A.M. Gorbachev, D.B. Radishev, A.L. Vikharev, M.A. Lobaev,
Contraction of microwave discharge in the reactor for chemical vapor deposition of
diamond, Tech. Phys. Lett. 45 (2019) 89–92, https://doi.org/10.1134/
s1063785019020032.
[12] K.F. Sergeichev, N.A. Lukina, A.P. Bolshakov, V.G. Ralchenko, N.R. Arutyunyan, I.
I. Vlasov, Polycrystal diamond growth in a microwave plasma torch, Plasma Phys.
Rep. 36 (2010) 1272–1277, https://doi.org/10.1134/s1063780x10130301.
[13] K.F. Sergeichev, N.A. Lukina, N.R. Arutyunyan, Atmospheric-pressure microwave
plasma torch for CVD technology of diamond synthesis, Plasma Phys. Rep. 45
(2019) 551–560, https://doi.org/10.1134/s1063780x19060096.
[14] J. Kim, H. Sakakita, H. Ohsaki, M. Katsurai, Microwave-excited atmospheric
pressure plasma jet with wide aperture for the synthesis of carbon nanomaterials,
Jpn. J. Appl. Phys. 54 (2015), https://doi.org/10.7567/jjap.54.01aa02.
[15] K.W. Hemawan, H. Gou, R.J. Hemley, Diamond synthesis at atmospheric pressure
by microwave capillary plasma chemical vapor deposition, Appl. Phys. Lett. 107
(2015), https://doi.org/10.1063/1.4934751.
[16] M. Naamoun, A. Tallaire, F. Silva, J. Achard, P. Doppelt, A. Gicquel, Etch-pit
formation mechanism induced on HPHT and CVD diamond single crystals by H2/
O2plasma etching treatment, Phys. Status Solidi A 209 (2012) 1715–1720, https://
doi.org/10.1002/pssa.201200069.
[17] S. Nad, Y. Gu, J. Asmussen, Growth strategies for large and high quality single
crystal diamond substrates, Diam. Relat. Mater. 60 (2015) 26–34, https://doi.org/
10.1016/j.diamond.2015.09.018.
[18] H. Yamada, A. Chayahara, Y. Mokuno, Method to increase the thickness and
quality of diamond layers using plasma chemical vapor deposition under (H, C, N,
O) system, Diam. Relat. Mater. 101 (2020), https://doi.org/10.1016/j.
diamond.2019.107652.
[19] E.V. Bushuev, V.Y. Yurov, A.P. Bolshakov, V.G. Ralchenko, E.E. Ashkinazi, A.
V. Ryabova, I.A. Antonova, P.V. Volkov, A.V. Goryunov, A.Y. Luk’yanov, Synthesis
of single crystal diamond by microwave plasma assisted chemical vapor deposition
with in situ low-coherence interferometric control of growth rate, Diam. Relat.
Mater. 66 (2016) 83–89, https://doi.org/10.1016/j.diamond.2016.03.023.
[20] E.V. Bushuev, V.Y. Yurov, A.P. Bolshakov, V.G. Ralchenko, A.A. Khomich, I.
A. Antonova, E.E. Ashkinazi, V.A. Shershulin, V.P. Pashinin, V.I. Konov, Express in
situ measurement of epitaxial CVD diamond flm growth kinetics, Diam. Relat.
Mater. 72 (2017) 61–70, https://doi.org/10.1016/j.diamond.2016.12.021.
[21] M.A. Lobaev, S.A. Bogdanov, D.B. Radishev, A.L. Vikharev, A.M. Gorbachev,
Method of power density determination in microwave discharge, sustained in
hydrogen–methane gas mixture, Diam. Relat. Mater. 66 (2016) 177–182, https://
doi.org/10.1016/j.diamond.2016.05.004.
[22] J. Ma, M.N.R. Ashfold, Y.A. Mankelevich, Validating optical emission spectroscopy
as a diagnostic of microwave activated CH4/Ar/H2 plasmas used for diamond
chemical vapor deposition, J. Appl. Phys. 105 (2009), 043302, https://doi.org/
10.1063/1.3078032.
[23] X. Duten, A. Rousseau, A. Gicquel, P. Leprince, Rotational temperature
measurements of excited and ground states of C2(d 3Πga3Πu) transition in a H2/
CH4 915 MHz microwave pulsed plasma, J. Appl. Phys. 86 (1999) 5299–5301,
https://doi.org/10.1063/1.371515.
[24] S.A. Bogdanov, A.M. Gorbachev, A.L. Vikharev, D.B. Radishev, M.A. Lobaev, Study
of microwave discharge at high power density conditions in diamond chemical
vapor deposition reactor by optical emission spectroscopy, Diam. Relat. Mater. 97
(2019), https://doi.org/10.1016/j.diamond.2019.04.030.
[25] A.L. Vikharev, A.M. Gorbachev, D.B. Radishev, Physics and application of gas
discharge in millimeter wave beams, J. Phys. D Appl. Phys. 52 (2019), https://doi.
org/10.1088/1361-6463/aae3a3.
[26] K.W. Hemawan, R.J. Hemley, Optical emission diagnostics of plasmas in chemical
vapor deposition of single-crystal diamond, J. Vac. Sci. Technol. A 33 (2015),
061302, https://doi.org/10.1116/1.4928031.
[27] K. Hassouni, F. Silva, A. Gicquel, Modelling of diamond deposition microwave
cavity generated plasmas, J. Phys. D Appl. Phys. 43 (2010), 153001, https://doi.
org/10.1088/0022-3727/43/15/153001.
[28] N. Derkaoui, C. Rond, K. Hassouni, A. Gicquel, Spectroscopic analysis of H2/CH4
microwave plasma and fast growth rate of diamond single crystal, J. Appl. Phys.
155 (2014), 233301, https://doi.org/10.1063/1.4883955.
[29] G. Shu, V.G. Ralchenko, A.P. Bolshakov, E.V. Zavedeev, A.A. Khomich, P.
A. Pivovarov, E.E. Ashkinazi, V.I. Konov, B. Dai, J. Han, J. Zhu, Evolution of
surface relief of epitaxial diamond flms upon growth resumption by microwave
plasma chemical vapor deposition, CrystEngComm 22 (2020) 2138–2146, https://
doi.org/10.1039/c9ce01933b.
[30] D.G. Goodwin, Scaling laws for diamond chemical vapor deposition. I. Diamond
surface chemistry, J. Appl. Phys. 74 (1993) 6888–6894, https://doi.org/10.1063/
1.355063.
[31] A. Charris, S. Nad, J. Asmussen, Exploring constant substrate temperature and
constant high pressure SCD growth using variable pocket holder depths, Diam.
Relat. Mater. 76 (2017) 58–67, https://doi.org/10.1016/j.diamond.2017.04.010.
[32] W. Müller-Sebert, E. W¨ orner, F. Fuchs, C. Wild, P. Koidl, Nitrogen induced increase
of growth rate in chemical vapor deposition of diamond, Appl. Phys. Lett. 68
(1996) 759–760, https://doi.org/10.1063/1.116733.
[33] Cs. Yan, Y.K. Vohra, Hk. Mao, R.J. Hemley, Very high growth rate chemical vapor
deposition of single-crystal diamond, Proc. Natl. Acad. Sci. U.S.A 99 (2002)
12523–12525, https://doi.org/10.1073/pnas.152464799.
[34] A. Chayahara, Y. Mokuno, Y. Horino, Y. Takasu, H. Kato, H. Yoshikawa,
N. Fujimori, The effect of nitrogen addition during high-rate homoepitaxial growth
of diamond by microwave plasma CVD, Diam. Relat. Mater. 13 (2004) 1954–1958,
https://doi.org/10.1016/j.diamond.2004.07.007.
[35] J. Achard, F. Silva, O. Brinza, A. Tallaire, A. Gicquel, Coupled effect of nitrogen
addition and surface temperature on the morphology and the kinetics of thick CVD
diamond single crystals, Diam. Relat. Mater. 16 (2007) 685–689, https://doi.org/
10.1016/j.diamond.2006.09.012.
[36] A.M. Zaitsev, Optical Properties of Diamond: a Data Handbook, Springer Science &
Business Media, 2001.
[37] L.-f. Hei, Y. Zhao, J.-j. Wei, J.-l. Liu, C.-m. Li, F.-x. Lü, Optical characterization of
single-crystal diamond grown by DC arc plasma jet CVD, Int. J. Miner. Metall.
Mater. 24 (2017) 1424–1430, https://doi.org/10.1007/s12613-017-1535-x.
[38] G. Shu, B. Dai, V.G. Ralchenko, A.A. Khomich, E.E. Ashkinazi, A.P. Bolshakov, S.
N. Bokova-Sirosh, K. Liu, J. Zhao, J. Han, J. Zhu, Epitaxial growth of mosaic
diamond: mapping of stress and defects in crystal junction with a confocal Raman
spectroscopy, J. Cryst. Growth 463 (2017) 19–26, https://doi.org/10.1016/j.
jcrysgro.2017.01.045.
[39] V. Yurov, E. Bushuev, A. Bolshakov, E. Ashkinazi, I. Antonova, E. Zavedeev,
A. Khomich, V. Voronov, V. Ralchenko, Etching kinetics of (100) single crystal
diamond surfaces in a hydrogen microwave plasma, studied with in situ low-
coherence interferometry, Phys. Status Solidi A 214 (2017), 1700177, https://doi.
org/10.1002/pssa.201700177.
A.P. Bolshakov et al.
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.
References (40)
- F. Silva, J. Achard, O. Brinza, X. Bonnin, K. Hassouni, A. Anthonis, K. De Corte, J. Barjon, High quality, large surface area, homoepitaxial MPACVD diamond growth, Diam. Relat. Mater. 18 (2009) 683-697, https://doi.org/10.1016/j. diamond.2009.01.038.
- M. Muehle, J. Asmussen, M.F. Becker, T. Schuelke, Extending microwave plasma assisted CVD SCD growth to pressures of 400 Torr, Diam. Relat. Mater. 79 (2017) 150-163, https://doi.org/10.1016/j.diamond.2017.09.013.
- Y. Mokuno, A. Chayahara, Y. Soda, H. Yamada, Y. Horino, N. Fujimori, High rate homoepitaxial growth of diamond by microwave plasma CVD with nitrogen addition, Diam. Relat. Mater. 15 (2006) 455-459, https://doi.org/10.1016/j. diamond.2005.11.046.
- Q. Liang, C.-s. Yan, Y. Meng, J. Lai, S. Krasnicki, H.-k. Mao, R.J. Hemley, Recent advances in high-growth rate single-crystal CVD diamond, Diam. Relat. Mater. 18 (2009) 698-703, https://doi.org/10.1016/j.diamond.2008.12.002.
- Y.J. Gu, J. Lu, T. Grotjohn, T. Schuelke, J. Asmussen, Microwave plasma reactor design for high pressure and high power density diamond synthesis, Diam. Relat. Mater. 24 (2012) 210-214, https://doi.org/10.1016/j.diamond.2012.01.026.
- A.P. Bolshakov, V.G. Ralchenko, V.Y. Yurov, A.F. Popovich, I.A. Antonova, A. A. Khomich, E.E. Ashkinazi, S.G. Ryzhkov, A.V. Vlasov, A.V. Khomich, High-rate growth of single crystal diamond in microwave plasma in CH4/H2 and CH4/H2/Ar gas mixtures in presence of intensive soot formation, Diam. Relat. Mater. 62 (2016) 49-57, https://doi.org/10.1016/j.diamond.2015.12.001.
- C.J. Widmann, W. Müller-Sebert, N. Lang, C.E. Nebel, Homoepitaxial growth of single crystalline CVD-diamond, Diam. Relat. Mater. 64 (2016) 1-7, https://doi. org/10.1016/j.diamond.2015.12.016.
- A.L. Vikharev, M.A. Lobaev, A.M. Gorbachev, D.B. Radishev, V.A. Isaev, S. A. Bogdanov, Investigation of homoepitaxial growth by microwave plasma CVD providing high growth rate and high quality of diamond simultaneously, Mater. Today Commun. 22 (2020), https://doi.org/10.1016/j.mtcomm.2019.100816.
- Q. Liang, C.Y. Chin, J. Lai, C.-s. Yan, Y. Meng, H.-k. Mao, R.J. Hemley, Enhanced growth of high quality single crystal diamond by microwave plasma assisted chemical vapor deposition at high gas pressures, Appl. Phys. Lett. 94 (2009), https://doi.org/10.1063/1.3072352.
- A.P. Bolshakov, V.G. Ralchenko, V.Y. Yurov, G. Shu, E.V. Bushuev, A.A. Khomich, E.E. Ashkinazi, D.N. Sovyk, I.A. Antonova, S.S. Savin, V.V. Voronov, M. Y. Shevchenko, B. Dai, J. Zhu, Enhanced deposition rate of polycrystalline CVD diamond at high microwave power densities, Diam. Relat. Mater. 97 (2019), https://doi.org/10.1016/j.diamond.2019.107466.
- S.A. Bogdanov, A.M. Gorbachev, D.B. Radishev, A.L. Vikharev, M.A. Lobaev, Contraction of microwave discharge in the reactor for chemical vapor deposition of diamond, Tech. Phys. Lett. 45 (2019) 89-92, https://doi.org/10.1134/ s1063785019020032.
- K.F. Sergeichev, N.A. Lukina, A.P. Bolshakov, V.G. Ralchenko, N.R. Arutyunyan, I. I. Vlasov, Polycrystal diamond growth in a microwave plasma torch, Plasma Phys. Rep. 36 (2010) 1272-1277, https://doi.org/10.1134/s1063780x10130301.
- K.F. Sergeichev, N.A. Lukina, N.R. Arutyunyan, Atmospheric-pressure microwave plasma torch for CVD technology of diamond synthesis, Plasma Phys. Rep. 45 (2019) 551-560, https://doi.org/10.1134/s1063780x19060096.
- J. Kim, H. Sakakita, H. Ohsaki, M. Katsurai, Microwave-excited atmospheric pressure plasma jet with wide aperture for the synthesis of carbon nanomaterials, Jpn. J. Appl. Phys. 54 (2015), https://doi.org/10.7567/jjap.54.01aa02.
- K.W. Hemawan, H. Gou, R.J. Hemley, Diamond synthesis at atmospheric pressure by microwave capillary plasma chemical vapor deposition, Appl. Phys. Lett. 107 (2015), https://doi.org/10.1063/1.4934751.
- M. Naamoun, A. Tallaire, F. Silva, J. Achard, P. Doppelt, A. Gicquel, Etch-pit formation mechanism induced on HPHT and CVD diamond single crystals by H2/ O2plasma etching treatment, Phys. Status Solidi A 209 (2012) 1715-1720, https:// doi.org/10.1002/pssa.201200069.
- S. Nad, Y. Gu, J. Asmussen, Growth strategies for large and high quality single crystal diamond substrates, Diam. Relat. Mater. 60 (2015) 26-34, https://doi.org/ 10.1016/j.diamond.2015.09.018.
- H. Yamada, A. Chayahara, Y. Mokuno, Method to increase the thickness and quality of diamond layers using plasma chemical vapor deposition under (H, C, N, O) system, Diam. Relat. Mater. 101 (2020), https://doi.org/10.1016/j. diamond.2019.107652.
- E.V. Bushuev, V.Y. Yurov, A.P. Bolshakov, V.G. Ralchenko, E.E. Ashkinazi, A. V. Ryabova, I.A. Antonova, P.V. Volkov, A.V. Goryunov, A.Y. Luk'yanov, Synthesis of single crystal diamond by microwave plasma assisted chemical vapor deposition with in situ low-coherence interferometric control of growth rate, Diam. Relat. Mater. 66 (2016) 83-89, https://doi.org/10.1016/j.diamond.2016.03.023.
- E.V. Bushuev, V.Y. Yurov, A.P. Bolshakov, V.G. Ralchenko, A.A. Khomich, I. A. Antonova, E.E. Ashkinazi, V.A. Shershulin, V.P. Pashinin, V.I. Konov, Express in situ measurement of epitaxial CVD diamond film growth kinetics, Diam. Relat. Mater. 72 (2017) 61-70, https://doi.org/10.1016/j.diamond.2016.12.021.
- M.A. Lobaev, S.A. Bogdanov, D.B. Radishev, A.L. Vikharev, A.M. Gorbachev, Method of power density determination in microwave discharge, sustained in hydrogen-methane gas mixture, Diam. Relat. Mater. 66 (2016) 177-182, https:// doi.org/10.1016/j.diamond.2016.05.004.
- J. Ma, M.N.R. Ashfold, Y.A. Mankelevich, Validating optical emission spectroscopy as a diagnostic of microwave activated CH4/Ar/H2 plasmas used for diamond chemical vapor deposition, J. Appl. Phys. 105 (2009), 043302, https://doi.org/ 10.1063/1.3078032.
- X. Duten, A. Rousseau, A. Gicquel, P. Leprince, Rotational temperature measurements of excited and ground states of C2(d 3Πg-a 3Πu) transition in a H2/ CH4 915 MHz microwave pulsed plasma, J. Appl. Phys. 86 (1999) 5299-5301, https://doi.org/10.1063/1.371515.
- S.A. Bogdanov, A.M. Gorbachev, A.L. Vikharev, D.B. Radishev, M.A. Lobaev, Study of microwave discharge at high power density conditions in diamond chemical vapor deposition reactor by optical emission spectroscopy, Diam. Relat. Mater. 97 (2019), https://doi.org/10.1016/j.diamond.2019.04.030.
- A.L. Vikharev, A.M. Gorbachev, D.B. Radishev, Physics and application of gas discharge in millimeter wave beams, J. Phys. D Appl. Phys. 52 (2019), https://doi. org/10.1088/1361-6463/aae3a3.
- K.W. Hemawan, R.J. Hemley, Optical emission diagnostics of plasmas in chemical vapor deposition of single-crystal diamond, J. Vac. Sci. Technol. A 33 (2015), 061302, https://doi.org/10.1116/1.4928031.
- K. Hassouni, F. Silva, A. Gicquel, Modelling of diamond deposition microwave cavity generated plasmas, J. Phys. D Appl. Phys. 43 (2010), 153001, https://doi. org/10.1088/0022-3727/43/15/153001.
- N. Derkaoui, C. Rond, K. Hassouni, A. Gicquel, Spectroscopic analysis of H2/CH4 microwave plasma and fast growth rate of diamond single crystal, J. Appl. Phys. 155 (2014), 233301, https://doi.org/10.1063/1.4883955.
- G. Shu, V.G. Ralchenko, A.P. Bolshakov, E.V. Zavedeev, A.A. Khomich, P. A. Pivovarov, E.E. Ashkinazi, V.I. Konov, B. Dai, J. Han, J. Zhu, Evolution of surface relief of epitaxial diamond films upon growth resumption by microwave plasma chemical vapor deposition, CrystEngComm 22 (2020) 2138-2146, https:// doi.org/10.1039/c9ce01933b.
- D.G. Goodwin, Scaling laws for diamond chemical vapor deposition. I. Diamond surface chemistry, J. Appl. Phys. 74 (1993) 6888-6894, https://doi.org/10.1063/ 1.355063.
- A. Charris, S. Nad, J. Asmussen, Exploring constant substrate temperature and constant high pressure SCD growth using variable pocket holder depths, Diam. Relat. Mater. 76 (2017) 58-67, https://doi.org/10.1016/j.diamond.2017.04.010.
- W. Müller-Sebert, E. Wörner, F. Fuchs, C. Wild, P. Koidl, Nitrogen induced increase of growth rate in chemical vapor deposition of diamond, Appl. Phys. Lett. 68 (1996) 759-760, https://doi.org/10.1063/1.116733.
- Cs. Yan, Y.K. Vohra, Hk. Mao, R.J. Hemley, Very high growth rate chemical vapor deposition of single-crystal diamond, Proc. Natl. Acad. Sci. U.S.A 99 (2002) 12523-12525, https://doi.org/10.1073/pnas.152464799.
- A. Chayahara, Y. Mokuno, Y. Horino, Y. Takasu, H. Kato, H. Yoshikawa, N. Fujimori, The effect of nitrogen addition during high-rate homoepitaxial growth of diamond by microwave plasma CVD, Diam. Relat. Mater. 13 (2004) 1954-1958, https://doi.org/10.1016/j.diamond.2004.07.007.
- J. Achard, F. Silva, O. Brinza, A. Tallaire, A. Gicquel, Coupled effect of nitrogen addition and surface temperature on the morphology and the kinetics of thick CVD diamond single crystals, Diam. Relat. Mater. 16 (2007) 685-689, https://doi.org/ 10.1016/j.diamond.2006.09.012.
- A.M. Zaitsev, Optical Properties of Diamond: a Data Handbook, Springer Science & Business Media, 2001.
- L.-f. Hei, Y. Zhao, J.-j.
- Wei, J.-l. Liu, C.-m. Li, F.-x. Lü, Optical characterization of single-crystal diamond grown by DC arc plasma jet CVD, Int. J. Miner. Metall. Mater. 24 (2017) 1424-1430, https://doi.org/10.1007/s12613-017-1535-x.
- G. Shu, B. Dai, V.G. Ralchenko, A.A. Khomich, E.E. Ashkinazi, A.P. Bolshakov, S. N. Bokova-Sirosh, K. Liu, J. Zhao, J. Han, J. Zhu, Epitaxial growth of mosaic diamond: mapping of stress and defects in crystal junction with a confocal Raman spectroscopy, J. Cryst. Growth 463 (2017) 19-26, https://doi.org/10.1016/j. jcrysgro.2017.01.045.
- V. Yurov, E. Bushuev, A. Bolshakov, E. Ashkinazi, I. Antonova, E. Zavedeev, A. Khomich, V. Voronov, V. Ralchenko, Etching kinetics of (100) single crystal diamond surfaces in a hydrogen microwave plasma, studied with in situ low- coherence interferometry, Phys. Status Solidi A 214 (2017), 1700177, https://doi. org/10.1002/pssa.201700177.
FAQs
AI
What is the maximum growth rate achieved for SCD at high pressures?
The study achieved a maximum single crystal diamond growth rate of 57 μm/h at 500 Torr.
How does plasma shape evolve with increasing pressure during synthesis?
Plasma shape transitions from spherical at 100 Torr to a vertically elongated ellipsoid at 600 Torr.
What are the implications of plasma volume reduction on growth rates?
Plasma volume decreases from 17 cm³ to 2 cm³ as pressure rises, influencing growth rate stability.
How does optical emission vary with pressure during diamond growth?
Bright emission lines for species such as C2 and CH increase with pressure, but saturate above 500 Torr.
What factors contribute to growth rate declines at very high pressures?
Growth rates decline due to reduced radical supply from compressed plasma and nitrogen contamination effects.