High pressure phase transformations in neodymium studied in a diamond anvil cell using diamond-coated rhenium gaskets

Article abstract of DOI:10.1088/0953-8984/11/34/303

Diamond anvil cells are used to generate high static pressures up to several megabars (hundreds of GPa) in very small volumes of material. We have explored a technique which employs a microwave plasma chemical vapour deposited diamond layer on one side of the rhenium gasket. The high yield strength of the diamond layer prevents excessive thickness reduction of the sample in the gasket hole. As a test case, we show energy dispersive x-ray diffraction data on rare earth metal neodymium to high pressures of 153 GPa using a synchrotron source. The increased sample thickness results in an unambiguous crystal structure determination of a monoclinic phase in neodymium above 75 GPa.

Full text of DOI:10.1088/0953-8984/11/34/303

Home Search Collections Journals About Contact us My IOPscience
High pressure phase transformations in neodymium studied in a diamond anvil cell using
diamond-coated rhenium gaskets
This content has been downloaded from IOPscience. Please scroll down to see the full text.
1999 J. Phys.: Condens. Matter 11 6515
(http://iopscience.iop.org/0953-8984/11/34/303)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 193.255.248.150
This content was downloaded on 30/01/2015 at 08:09
Please note that terms and conditions apply.

J. Phys.: Condens. Matter 11 (1999) 6515–6520. Printed in the UK
PII: S0953-8984(99)03374-3
High pressure phase transformations in neodymium studied in
a diamond anvil cell using diamond-coated rhenium gaskets
Jagannadham Akella†, Samuel T Weir†, Yogesh K Vohra‡, Heather Prokop‡,
Shane A Catledge‡ and Gary N Chesnut‡
†
‡
Lawrence Livermore National Laboratory (LLNL), Livermore, CA 94550, USA
Department of Physics, University of Alabama at Birmingham (UAB), Birmingham,
AL 35294-1170, USA
Received 13 April 1999
Abstract. Diamond anvil cells are used to generate high static pressures up to several megabars
(hundreds of GPa) in very small volumes of material. We have explored a technique which employs
a microwave plasma chemical vapour deposited diamond layer on one side of the rhenium gasket.
The high yield strength of the diamond layer prevents excessive thickness reduction of the sample
in the gasket hole. As a test case, we show energy dispersive x-ray diffraction data on rare
earth metal neodymium to high pressures of 153 GPa using a synchrotron source. The increased
sample thickness results in an unambiguous crystal structure determination of a monoclinic phase
in neodymium above 75 GPa.
Single crystal diamonds and metal gaskets are the two most important components in a modern
diamond anvil cell (DAC) designed for research in the multi-megabar (100–1000 GPa) pressure
range. Single crystal diamond anvils are carefully selected with low internal strains, low
impurity level concentrations and high crystal perfection [1]. Static pressures of over 300 GPa
can be reached in DAC devices, which employ the use of single crystal diamond anvils with
bevelled tips. The sample of interest is placed within a small through-hole in a metal gasket,
which is pressed between the two anvils. During pressurization, the gasket plastically flows
around the tips of the opposing anvils and acts to maintain a uniform stress distribution within
the sample, as well as to hold the sample in place. With increasing pressure, however, the
sample thickness is reduced due to the large deformation of the gasket itself. For the case
of a gasket compressed in a DAC in an axially symmetric manner between flat rigid anvils
with a no-slip boundary condition between the gasket and anvils, the following relation holds
between the gasket flow stress σ, the sample thickness h and the pressure distribution dP/dr.
σ = h(dP/dr).
(1)
If the sample thickness, h, within the gasket hole becomes too small, the reduced sample
volume degrades the x-ray diffraction signals as well as optical signals generated using laser
excitation.
Equation (1) has been employed to measure the flow stress of rhenium at high pressure [2]
and more recently to measure the flow stress of uranium and tantalum at ultra-high pressure [3].
Recent finite-element modelling of diamond deformation at multi-megabar pressure [4] has
under-scored the need for high yield strength gasket materials for optimizing sample thickness
at high pressures. A diamond–metal composite gasket is one possible way to improve the
sample thickness by utilizing the high yield strength of diamond and the plastic deformation
of the metallic component.
0
953-8984/99/346515+06$30.00 © 1999 IOP Publishing Ltd
6515

6516
J Akella et al
Figure 1. Optical micrograph of the diamond-coated rhenium gasket. The microcrystalline
diamond film thickness is approximately 10 microns. The sample hole is 50 microns in diameter.
This gasket was successfully tested to pressures of 153 GPa in a diamond anvil cell.
To address this problem, microwave plasma-enhanced chemical vapour deposition (CVD)
was used to deposit a layer of polycrystalline diamond onto a rhenium gasket. The high yield
strength of the diamond layer prevents excessive thickness reduction of the sample in the gasket
hole. Rhenium is chosen for the gasket material because of its suitability for high pressure/high
temperature measurements and earlier high pressure studies have shown that the ambient
pressure hcp phase is stable to 216 GPa [5]. While CVD diamond deposition is routinely
carried out on materials such as silicon (for electronic-based applications) as well as on several
metals such as tungsten carbide and titanium alloys (for wear-resistant applications), diamond
coating of rhenium has not, to the authors’ knowledge, been published. One concern involving
the vapour deposition of diamond on metals is the residual film stress caused by the significant
thermal mismatch, which can lead to film delamination. For example, a residual stress of up
to 7.1 GPa has been reported [6] for diamond-coated Ti–6Al–4V (a typical aerospace alloy).
For rhenium, a residual thermal stress of about 5 GPa (compressive) is still expected, assuming
a typical substrate deposition temperature of 1173 K. Despite the thermal stress expected in
this coating, we have deposited a thick, well adhered diamond layer on the rhenium gasket
and used it in high pressure experiments to 153 GPa on rare earth metal neodymium. The
success in coating rhenium with diamond and in obtaining high pressures in the DAC may
be related to the unusually high melting temperature (3459 K) of rhenium. The diamond
deposition temperature (1173 K) is well below half of its melting temperature, at which point
dislocations would normally begin to anneal out of the material and reduce the work-hardening
effects induced during the gasket indentation process. Therefore, the CVD diamond coating
process on rhenium would not degrade the work hardening of rhenium achieved during the
pre-indentation process in a diamond anvil cell.
◦
A standard diamond anvil cell using bevel diamonds of 70 micron central flat, 7 bevel, and
300 micron culet size was employed in the present experiments. Before depositing the diamond
film onto the gasket, the gasket was indented in the DAC and a 50 micron diameter sample hole
was drilled according to standard procedures. Figure 1 shows an optical micrograph of the

Diamond-coated rhenium gaskets in DAC
6517
Figure 2. Glancing angle x-ray diffraction pattern (copper Kα radiation, λ = 0.1542 nm) from
the diamond-coated hexagonal close-packed rhenium gasket at ambient pressure before placement
in a diamond anvil cell. Diffraction peaks from both rhenium and diamond are labelled.
approximately 10 micron thick diamond layer on rhenium grown in a 6 kW microwave plasma
CVD system. This coating was grown on one side of the gasket only at a processing pressure of
90 Torr, a methane fraction of 2% (in hydrogen), a microwave power of 1 kW and a substrate
temperature of 1173 K. The deposition time was 6 hours. The side of the rhenium gasket
exposed to plasma was covered with a homogenous and well adhered diamond layer. The film
morphology shows well faceted, micron-size grains, which are of high diamond phase purity
as determined by Raman spectroscopy. Figure 2 shows the glancing-angle x-ray diffraction
spectrum of the diamond-coated rhenium gasket. The spectrum was indexed to a mixture of
cubic diamond phase and a hexagonal close-packed phase of rhenium metal. The measured
lattice parameters of rhenium are a = 0.2772 nm and c = 0.4447 nm, which are within
0
0
.4% of the literature values. The measured lattice parameter of the cubic diamond structure is
.3564 nm, which is within 0.1% of the literature value. Our attempts to deposit diamond layers
on both sides of the gasket were not completely successful due to graphitization of the diamond
layer not exposed to plasma. Therefore, all high pressure experiments were carried out using
gasket coated with a diamond layer on one side only. Also, we expect some diamond coating
of the inside walls of the gasket hole. However no cross-sectional studies on the gasket were
performed to examine the details of diamond coating on the inside walls of the sample chamber.
We have chosen the rare-earth metal neodymium as a test case sample for the applicability
of the diamond-coated rhenium gasket. Copper was employed as an internal pressure standard
in our x-ray diffraction study. Phase transformations induced by ultra-high pressures in rare-
earth metals are an active field of study at the present time [7, 8]. Complex crystal structures

6518
J Akella et al
Figure 3. The energy dispersive x-ray diffraction spectrum from the neodymium (Nd) sample and
a copper (Cu) pressure marker contained in a diamond-coated rhenium gasket at 89 GPa. The Cu
∗
diffraction peaks marked by are labelled according to an fcc phase. The Nd diffraction peaks
are indexed to space group C2/m with a = 0.4858 ± 0.0004 nm, b = 0.2712 ± 0.0002 nm,
◦
c = 0.5051 ± 0.0004 nm and β = 118.6 ± 0.2 . The fluorescence lines from Nd, Kα and Kβ are
also labelled. e is an escape peak from the germanium detector.
are observed in rare-earth metals at high pressures and identification of these phases is limited
by the quality of the x-ray diffraction patterns due to thin samples (few microns) at megabar
pressures. Neodymium crystallizes in the double hexagonal close-packed (dhcp) phase at
ambient pressures. The following sequence of transformations is known for neodymium
[9, 10] to 60 GPa:
3
.8 GPa
18 GPa
40 GPa
dhcp −→ fcc −→ distorted fcc −→ hp3.

Diamond-coated rhenium gaskets in DAC
6519
Table 1. Comparison of the observed and calculated interplanar spacings dhkl and the observed
and calculated intensities for the energy dispersive x-ray diffraction data for neodymium at 89 GPa.
The calculated intensities are based on a monoclinic C2/m phase described in the text.
hkl
dobs (Å)
dcalc (Å)
Iobs
Icalc
¯
1
0
2
1
1
3
2
2
1
0
3
4
2
1
4
1
11
02
2.287
2.218
2.128
1.850
1.418
1.371
1.262
1.185
1.116
1.107
1.093
1.062
0.937
0.909
0.886
0.852
2.287
2.218
2.128
1.851
1.417
1.369
1.262
1.186
1.114
1.105
1.092
1.059
0.936
0.908
0.886
0.853
100
32
51
31
8
100
47
41
55
20
3
¯
02
11
¯
13
¯
12
10
12
4
¯
04
5
¯
21
1
5
13
4
¯
04
1
2
¯
14
6
1
¯
04
4
2
2
1
21
¯
15
2
1
¯
21
7
2
1
1
31
The ultra-high pressure behaviour of neodymium above 60 GPa is of interest to investigate the
f-shell delocalization phenomenon and related phase transformations. We have investigated
high pressure transformations in neodymium to 153 GPa in a DAC using a diamond-coated
rhenium gasket.
After the gasket was coated with diamond, a neodymium sample along with a copper
pressure marker was placed in the gasket hole. Energy dispersive x-ray diffraction
measurements were carried out at beam-line X-17C, NSLS, Brookhaven National Laboratory.
A micro-collimated x-ray beam of 11 µm × 6 µm was employed in our x-ray diffraction
measurements. In our experiments, the sequence of transformations observed below 60 GPa is
similar to earlier results [9, 10] as documented above. We observed a new phase transformation
in neodymium at a pressure of 75 ± 5 GPa from a hexagonal phase with three atoms/cell,
hp3 phase [9], to a monoclinic C2/m phase [11]. Figure 3 shows the energy dispersive
x-ray diffraction spectrum from the neodymium sample and copper pressure marker contained
in the diamond-coated rhenium gasket at 89 GPa. The pressures are calculated from the
equation of state of copper based on the procedures described in [5] and [6]. The sample is
indexed to the space group C2/m (monoclinic phase with four atoms/cell) as described in
[
9] with a = 0.4858 ± 0.0004 nm, b = 0.2712 ± 0.0002 nm, c = 0.5051 ± 0.0004 nm
◦
and β = 118.6 ± 0.2 . The four atoms occupy the 4(i) positions of C2/m with coordinates
(
0.280, 0, 0.252) similar to values determined for cerium [11]. Table 1 shows a comparison of
the observed and calculated interplanar spacing dhkl and the observed and calculated intensities
for the x-ray diffraction data for neodymium at 89 GPa. It can be seen from table 1 that the
C2/m phase gives an excellent fit to the observed 17 diffraction peaks. The large numbers of
observable diffraction peaks are due to a thicker sample afforded by the use of a diamond-coated
rhenium gasket. The measured compression of neodymium at 89 GPa is V/V0 = 0.428. The
C2/m phase is stable to the highest pressures of 153 GPa (volume compression V/V0 = 0.37).
The diamond layer on top of the rhenium metal did not crack to the highest pressure in these
experiments.

6520
J Akella et al
We offer the following conclusions:
(1) Thick diamond films can be successfully deposited on rhenium metal using the standard
CH4/H2 chemistry by a microwave plasma CVD process.
(
2) High pressure x-ray diffraction studies on a neodymium sample were successfully carried
out to 153 GPa in a diamond anvil cell with excellent diffraction pattern quality. We
attribute the excellent diffraction pattern quality to a thicker sample afforded by the use
of a diamond-coated rhenium gasket.
(
3) The high quality x-ray diffraction patterns resulted in a positive identification of a C2/m
phase in compressed neodymium above 75 ± 5 GPa. This C2/m phase is similar to the
one observed in compressed cerium. This C2/m phase of neodymium was found to be
stable to pressures as high as 153 GPa.
Acknowledgments
Jagannadham Akella and Samuel T Weir acknowledge support by the B-division at the
Lawrence Livermore National Laboratory. Yogesh K Vohra, Shane A Catledge and Gary
N Chesnut acknowledge support from National Science Foundation (NSF) grant No DMR-
9704428. Heather Prokop acknowledges support from the NSF Research Experience for
Undergraduates (REU) site grant under DMR-9619405.
References
[
[
[
[
[
[
[
[
[
1] Liu J and Vohra Y K 1996 Appl. Phys. Lett. 68 2049
2] Jeanloz R, Godwal B K and Meade C 1991 Nature 349 687
3] Weir S T, Akella J, Ruddle C, Goodwin T and Hsuing L 1998 Phys. Rev. B 58 11 258
4] Merkel S, Hemley R J and Mao H K 1999 Appl. Phys. Lett. 74 656
5] Vohra Y K, Duclos S J and Ruoff A L 1987 Phys. Rev. B 36 9790
6] Ager J W III and Drory M D 1993 Phys. Rev. B 48 2601
7] Chesnut G N and Vohra Y K 1998 Phys. Rev. B 57 10 221
8] Chesnut G N and Vohra Y K 1999 Phys. Rev. Lett. 82 1712
9] Zhao Y C, Porsch F and Holzapfel W B 1994 Phys. Rev. B 50 6603
[
[
10] Akella J, Xu J and Smith G S1986 Physica B 139/140 285
11] McMahon M I and Nelmes R J 1997 Phys. Rev. Lett. 78 3884