Diamond & Related Materials 147 (2024) 111236

Contents lists available at ScienceDirect

Diamond & Related Materials

journal homepage: www.elsevier.com/locate/diamond

Fabrication of lateral diamond MOSFET with buried pn-junctions by

diamond surface planarization based on carbon solid solution into nickel
Tsubasa Kano a , Kimiyoshi Ichikawa b , Kan Hayashi a, b , Taro Yoshikawa b, c , Takao Inokuma a ,
Satoshi Yamasaki b , Norio Tokuda a, b , Tsubasa Matsumoto a, b, *
a
Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, Ishikawa 920-1192, Japan
b
Nanomaterials Research Institute, Kanazawa University, Kanazawa, Ishikawa 920-1192, Japan
c
Innovation and Business Development Headquarters, Dicel Corporation, Kakuma-machi, Kanazawa, Ishikawa, 920-1192, Japan

A R T I C L E I N F O A B S T R A C T

Keywords: New fabrication processes for buried p+-type diamond using etching techniques based on the solid-solution
Diamond reaction of carbon with nickel were proposed and demonstrated. Specifically, an inversion-channel diamond
Solid-solution reaction metal-oxide-semiconductor field-effect transistors (MOSFETs) with source and drain regions buried in the body
Surface planarization
were fabricated, and their operation were confirmed. An etching process using etching techniques based on the
MOSFET fabrication
solid-solution reaction of carbon with nickel was applied to realize trench formation and planarization after the
growth of a p+-type diamond. The fabricated MOSFET exhibited drain current density equivalent to that of
conventional MOSFET with source and drain p+-type regions deposited as islands on an n-type body, and ideal
field-effect transistor characteristics with linear and saturation regions, and the drain current density was
controlled by the gate voltage. However, the surface protrusion structure owing to the incomplete planarization
process limited the device characteristics. The proposed fabrication processes for buried layers are expected to
function as an important selective doping technique for diamond electronic devices such as ion implantation.

1. Introduction Recently, diamond etching using the solid-solution reaction of car­

bon with nickel (Ni) was investigated as a new etching technique
Diamond is a promising material for power electronics owing to their [6–13]. When diamond and Ni are brought into contact at high tem­
high breakdown electric field [1], high thermal conductivity [2], and peratures, the carbon atoms at the Ni/diamond interface dissolve within
high electron and hole mobilities [3]. However, their physical hardness Ni and the diamond is etched. As this process is a pure interfacial re­
and tolerance to chemical processes hinder the fabrication of electric action, it is expected to be an ideal etching method because, unlike
device structures. For example, the wet etching process utilized in the plasma etching, it does not induce any damage near the interface.
etching of semiconductors such as silicon is not suitable for diamond. Therefore, studies have focused on etching mechanisms and the control
Consequently, plasma etching processes such as inductively coupled of etching rates via temperature and gas atmospheres [14–16]. The
plasma are used as an alternative technology; however, plasma damage etching rate by the solid-solution reaction of carbon increases with
induced near the surface after etching degrades the characteristics of the increasing temperature; for example, it can reach 30 μm/h at 1200 ◦ C in
fabricated device [4]. In addition, selective doping techniques such as a hydrogen atmosphere [10]. Furthermore, a hydrogen atmosphere can
ion implantation are essential for the fabrication of semiconductor de­ maintain the carbon unsaturated state of Ni and facilitate fast and
vices. However, the ion implantation of diamond is difficult because continuous etching [14]. Specifically, hydrogen molecules decompose
heat treatment to recover the defects created by ion implantation results into hydrogen atoms owing to the catalytic activity of Ni. When the
in a phase transition from diamond to graphite, the most stable structure solid-dissolved carbon atoms at the Ni/diamond interface reach the gas-
of solid carbon [5]. Therefore, improvements in the annealing process or phase/Ni interface, they combine with the hydrogen atoms and desorb
the development of an alternative technology for ion implantation is from Ni into the gas phase as methane gas. This carbon desorption
important for facilitating the widespread use of diamond devices. maintains the carbon unsaturated state of Ni, thereby maintaining a high

* Corresponding author at: Nanomaterials Research Institute, Kanazawa University, Kanazawa, Ishikawa 920-1192, Japan.
E-mail address: t-matsumoto@se.kanazawa-u.ac.jp (T. Matsumoto).

https://doi.org/10.1016/j.diamond.2024.111236
Received 7 March 2024; Received in revised form 16 May 2024; Accepted 23 May 2024
Available online 30 May 2024
0925-9635/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
T. Kano et al. Diamond & Related Materials 147 (2024) 111236

etching rate. Graphite precipitation at the Ni/diamond interface after fabricated on the sample using photolithography to determine the Ni
etching has been reported [17]. Despite studies on the etching mecha­ deposition area. Thereafter, Ni was deposited via vacuum evaporation,
nism and interface state, their application in device fabrication is yet to and the Ni films were patterned using a lift-off process (Fig. 1(a)). The
be reported. Thus, this study proposed a new device fabrication process thickness of the Ni films was approximately 400 nm. The sample was
using etching by the solid-solution reaction of carbon with Ni as the next then annealed (4 % H2/Ar atmosphere, flow rate 1 L/min, 900 ◦ C, 2 h) to
step. form trench structures via etching using a solid-solution reaction of
Conventional inversion p-channel diamond metal-oxide- carbon with Ni films. Subsequently, the sample was immersed in an
semiconductor field-effect transistors (MOSFETs) have the p+-type H2SO4/HNO3 (3:1) mixed acid at 220 ◦ C for 30 min to remove the Ni
source and drain (S/D) islands onto the n-type body without trench films (Fig. 1(b)). The trenched sample was then subjected to hydrogen
structure, due to the selective S/D growth using a Ti/Au metal mask plasma treatment using MPCVD under the following conditions. The
[18–20]. However, this structure hinders the understanding and the microwave power, pressure, H2 flow rate, and time were 600 W, 30 kPa,
improvement of the device characteristics, including the high contact 100 sccm, and 2 h, respectively. The sample was then immersed in a
resistance between the inversion channel and the p+-type region. H2O2/H2SO4 (2:1) piranha solution at 200 ◦ C for 10 min to remove
Consequently, a structure wherein the p+-type S/D regions are buried in contamination before p+-type layer growth. After cleaning, a heavily
an n-type body must be fabricated. In recent years, a method has been boron-doped p+-type layer was deposited on the entire surface of the
reported in which a buried structure is formed by performing dry sample following hydrogen plasma treatment using an MPCVD system
etching before selective growth [21], and a method in which a buried (Fig. 1(c)). The hydrogen plasma was sufficiently weak to prevent the
layer is formed by lateral growth without a mask in a trench structure introduction of defects during the process. Herein, a polycrystalline Ni
formed by dry etching [22]. However, they require very advanced substrate (4 mm × 4 mm × 0.8 mm, Crystal Base Co., Ltd) with a mirror
techniques, such as dry etching and subsequent plasma CVD growth- polished surface was ultrasonically cleaned and annealed (4 % H2/Ar
resistant mask techniques, and nucleation-free lateral growth tech­ atmosphere, flow rate 1 L/min, 1200 ◦ C, 1 h) to remove a natural oxide
niques. Thus, this study fabricated a MOSFET structure with buried S/D film and improve its crystallinity. Consequently, the sample brought in
regions using a new and simple device fabrication process and demon­ contact with this Ni substrate was annealed (4 % H2/Ar atmosphere, 1 L/
strated its operation. Finally, we discussed the challenges of new device min, 800 ◦ C, 2 h) to remove unwanted surface protuberances on the
fabrication processes. sample surface through solid-solution reaction of carbon with the Ni
substrate (Fig. 1(d)). Since the planarization process for the (111) plane
2. Experimental requires more precise etching, the annealing temperature was set lower
than that in the previous study for the (100) plane to reduce the etching
Fig. 1 shows an overview of the fabrication process of the buried rate [9]. Differences in surface orientation were not considered.
structure. Two types of carbon solid-solution reactions with Ni were Consequently, the sample was additionally annealed with steep tem­
used in this process: 1) trench formation etching using thin Ni films [8] perature change (4 % H2/Ar atmosphere, flow rate 1 L/min, 10 cycles
and 2) planarization etching using bulk Ni [9]. alternating 700 and 100 ◦ C) to facilitate uniform deposition of at the
A high-pressure and high-temperature (HPHT) synthesis Ib-type diamond/Ni substrate interface to detach the diamond sample from the
single-crystal diamond (111) substrate (2 mm × 2 mm × 0.3 mm) was Ni substrate. Any remaining graphite on the sample surface was
used. First, the substrate was immersed in a H2O2/H2SO4 (2:1) piranha removed through immersion in a H2SO4/HNO3 (3:1) mixed acid at
solution at 200 ◦ C for 10 min to remove surface contamination. There­ 220 ◦ C for 30 min (Fig. 1(e)). Finally, the MOSFET structure was
after, a nitrogen (N)-doped n-type diamond body was deposited on the fabricated by depositing Al2O3 with thickness of 50 nm on top using an
substrate using a microwave plasma chemical vapor deposition ALD system and forming electrodes via photolithography.
(MPCVD) system (Alios, Inc.) [20]. Subsequently, the sample was The surface morphologies of the samples were observed using a laser
immersed in a H2SO4/HNO3 (3:1) mixed acid at 220 ◦ C for 20 min to microscope (LM, OLYMPUS, LEXT OLS4100). Electrical characteristics
remove surface contamination. After cleaning, a photoresist mask was were measured at room temperature (RT) in air using a semiconductor
parameter analyzer (Keithley 4200-SCS).

3. Results and discussion

Fig. 2 shows the (a) optical image, (b) cross-sectional schematic of

the red dashed line area, and (c) drain current density (Id) vs. drain
voltage (Vds) characteristics at RT of an inversion p-channel diamond
MOSFET with Lg = 9 μm and Wg = 50 μm with buried S/D structure.
Gate voltage (Vgs) and Vds were applied in the range of 0 to − 12 and 0 to
− 7 V, respectively, with voltage steps of − 2 and − 0.1 V, respectively.
Fig. 2(c) indicates that this MOSFET could control Id with the Vgs and
exhibited ideal MOSFET characteristics with linear and saturation re­
gions. These results indicate that a process using etching techniques
involving a solid-solution reaction of carbon with Ni can be applied to
the fabrication process of a device structure that forms p+-type S/D re­
gions buried in diamond. The root-mean-square roughness (Sq) of the
sample surface in Fig. 2(a) was 110 nm, which is more than 100 times
rougher than that of a conventional MOSFET [19]. Nevertheless, the
maximum Id was − 1.6 mA/mm, which is equivalent to − 1.7 mA/mm of
the conventional MOSFET with S/D p+-type regions deposited as islands
on an n-type body [20]. To understand the surface roughness and
characteristics, the geometry of the diamond structure for each process
is discussed.
Fig. 3 shows the two-dimensional (2D)-LM height images and cross-
Fig. 1. Overview of the fabrication process for buried p+-type S/D regions. sectional profiles of the sample surface after the formation of the trench

2
T. Kano et al. Diamond & Related Materials 147 (2024) 111236

Fig. 2. (a) Optical image, (b) its cross-sectional profiles of red dashed line area of (a), and (c) drain current density (Id) vs. drain voltage (Vds) characteristics at RT of
the MOSFET with buried S/D structure. The lines at Vgs = 0, − 2, and − 4 V overlap.

However, the roughness of the etched surface increased from less than
10 nm to 110 nm as shown in Figs. 4 (a) and (b). The cause of the
roughness is not known. Studies are underway to consider step bunching
due to the diamond’s off-angle and expansion of Ni due to solute carbon
as one of the causes. Owing to the formation of the inversion channel at
the top of the surface protrusion structure, typically, contact between
the inversion channel and the buried S/D regions is difficult. However,
as shown in Fig. 4(a), a thin p+-type layer was deposited on the side of
the surface protrusion structure, which facilitated the connection of the
inversion p-channel and buried S/D regions, and the MOSFET func­
tioned appropriately.
Fig. 5 shows the maximum Id as a function of the height of the surface
protrusion structure for the channel forming region in MOSFETs, with
Wg = 50 μm, that were confirmed to operate. “The surface protrusion
structure height” is the difference between the highest and lowest por­
tions in the cross section of the structure formed in the gate region.
Fig. 3. 2D-LM height images of the sample surface after (a) trench structures Herein, Id decreased with increase in the height of the surface protrusion
formation and (b) hydrogen plasma treatment. (c) and (d) are cross-sectional structure. It is possible that the distance between the source and drain,
profiles corresponding to the red lines in (a) and (b), respectively. that is, the channel length, has simply become longer. It is also presumed
that a thin layer was deposited on the sidewall of the surface protrusion
structures and hydrogen plasma treatment. There is a Tilt direction at structure during the deposition of the p+-type layer, as shown in the
the top of the image. Significant roughness was observed at the bottom cross-sectional profiles in Fig. 5. The growth conditions of the p+-type
of the trench structure that was formed by etching with the Ni thin films layer in this study were optimized for the (111) surface, and the sidewall
(Fig. 3(a)). This roughness may be attributed to the etching conditions has a surface orientation other than (111), so the boron concentration is
not being optimal. Irregular lamellar structures are known to form at the likely to be low. Consequently, Id flows from the source region through
bottom of the pit owing to the incomplete etching of the diamond (111) the lightly B-doped layer on the sidewall to the inversion p-channel and
surface using a solid-solution reaction of carbon [23]. In this etching, then again through the lightly B-doped layer on the sidewall to the drain
lamellar structures were formed on the trench bottom surface owing to region. The thin lightly B-doped layer functioned as a resistor, which
non-optimal conditions, resulting in roughness. This is thought to be due was considered as the primary cause of the limitations of MOSFETs
to crystalline anisotropy originating from the characteristics of the (111) characteristics. In fact, the cross-sectional profile in Fig. 5 indicates that
plane, wherein one carbon atom has three covalent bonds, but the for­ the MOSFET with a low Id has an obvious surface protrusion structure
mation mechanism is not known. Fig. 3(b) shows that 2 h of the and the sidewall layer remains. Thus, the region between the S/D of the
hydrogen plasma treatment removed the unevenness at the bottom of MOSFET exhibited a high resistance. Therefore, the most important
the trench and the Sq value decreased from 130 to 50 nm. Furthermore, issue in this process was the elimination of the surface protrusion
the sidewalls of the trench became thinner because of the horizontal structure below the gate via planarization to ensure that the current did
etching, and the trench size expanded. A comparison of Figs. 3(a) and 3 not flow through the high-resistance sidewall layer.
(b) revealed that the trench size widened by 3 and 13 μm vertically and
horizontally, respectively. 4. Conclusion
Fig. 4 shows the 2D-LM height images and cross-sectional profiles of
the sample surface after (a) deposition of the p+-type layer and (b) This study fabricated a MOSFET structure with buried p+-type S/D
planarization. After the planarization process, the area around the regions using a new process involving etching techniques based on the
trench, including the channel formation region, was etched from a solid-solution reaction of carbon with Ni. Consequently, its electrical
height of 800 nm to height of 20 nm, as shown in Figs.4 (c) and (d). characteristics were evaluated. Based on observations of ideal MOSFETs

3
T. Kano et al. Diamond & Related Materials 147 (2024) 111236

Fig. 4. 2D-LM height images of the sample surface after (a) p+-layer deposition and (b) planarization. (c) and (d) are cross-sectional profiles corresponding to the red
line areas in (a) and (b), respectively. The boundary of p+-n interface was predicted from the surface morphology in the previous process.

development of diamond power electronics.

CRediT authorship contribution statement

Kimiyoshi Ichikawa: Validation, Data curation. Kan Hayashi:

Validation, Data curation. Taro Yoshikawa: Validation, Data curation.
Takao Inokuma: Supervision, Resources. Satoshi Yamasaki: Writing –
review & editing, Supervision. Norio Tokuda: Visualization, Supervi­
sion, Resources, Project administration, Methodology, Funding acqui­
sition. Tsubasa Matsumoto: Writing – original draft, Visualization,
Validation, Investigation, Formal analysis, Data curation,
Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial

interests or personal relationships that could have appeared to influence
the work reported in this paper.
Fig. 5. Maximum Id as a function of the height of surface protrusion structure
Data availability
for channel formation region. Cross-sectional profiles that are expected from
the profiles after hydrogen plasma and p+-layer growth show the surface pro­
trusion structure below the gate of MOSFETs pointed by red arrows. Data will be made available on request.

operation, we demonstrated that this process offered the potential to Acknowledgements

replace ion implantation technology. Furthermore, comparisons of
MOSFET characteristics and diamond structures in each process clarified This work was partially supported by the NEDO Feasibility Study
that the most important issue in this process is roughness after etching, Program (Uncharted Territory Challenge 2050) under Grant No.
especially the remaining surface protrusion structure in the channel 19101600-0, the JST FOREST Program (Grant No. JPMJFR20353078,
formation region. The establishment of a condition for elimination of the Japan), the Kanazawa University CHOZEN Project 2022, and the JSPS
surface protrusion structure contributes improvement of MOSFET KAKENHI under Grant Number JP21H01363.
characteristics. The etching process using a solid-solution reaction of
carbon with Ni, which has high selectivity where only the contact area References
between Ni and diamond is etched, facilitated the fabrication of more
[1] M.I. Landstrass, M.A. Plano, M.A. Moreno, M.C. Williams, L.S. Pan, D.R. Kania,
complex structures that are difficult to fabricate using conventional S. Han, Device properties of homoepitaxially grown diamond, Diam. Relat. Mater.
methods. Thus, the proposed method is expected to contribute to the 2 (1993) 1003–1037, https://doi.org/10.1016/0925-9635(93)90269-8.

4
T. Kano et al. Diamond & Related Materials 147 (2024) 111236

[2] Y. Yamamoto, T. Imai, K. Tanabe, T. Tsuno, Y. Kumazawa, N. Fujimori, The [13] H.-A. Mehedi, J.-C. Arnault, D. Eon, C. Hebert, D. Carole, F. Omnes, E. Gheeraert,
measurement of thermal properties of diamond, Diam. Relat. Mater. 6 (1997) Etching mechanism of diamond by Ni nanoparticles for fabrication of nanopores,
1057–1061, https://doi.org/10.1016/S0925-9635(96)00772-8. Carbon 59 (2013) 448–456, https://doi.org/10.1016/j.carbon.2013.03.038.
[3] I. Akimoto, Y. Handa, K. Fukai, N. Naka, High carrier mobility in ultrapure [14] C. Schreyvogel, S. Temgoua, C. Giese, V. Cimalla, J. Barjon, C.E. Nebel, Fabrication
diamond measured by time-resolved cyclotron resonance, Appl. Phys. Lett. 105 of n-type doped v-shaped structures on (100) diamond, Phys. Status Solidi A 218
(2014) 032102, https://doi.org/10.1063/1.4891039. (2021) 2000502, https://doi.org/10.1002/pssa.202000502.
[4] K. Eriguchi, K. Ono, Quantitative and comparative characterizations of plasma [15] S. Tulic, T. Waitz, M. Caplovicova, G. Habler, V. Vretenar, T. Susi, V. Skakalova,
process-induced damage in advanced metal-oxide-semiconductor devices, J. Phys. Catalytic graphitization of single-crystal diamond, Carbon 185 (2021) 300–313,
D. Appl. Phys. 41 (2008) 024002, https://doi.org/10.1088/0022-3727/41/2/ https://doi.org/10.1016/j.carbon.2021.08.082.
024002. [16] Y. Li, Y. Kim, P.V. Bakharev, W.K. Seong, C. Hyun, D.C. Camacho-Mojica, L. Zhang,
[5] P.-H. Wu, W.-H. Ku, K.-A. Chiu, L. Chang, Radiation damage in (001) diamond B.V. Cunning, T.J. Shin, G. Lee, R.S. Ruoff, Dissolving diamond: kinetics of the
induced by phosphorus ion implantation, Phys. Status Solidi A 219 (2022) dissolution of (100) and (110) single crystals in nickel and cobalt films, Chem.
2100829, https://doi.org/10.1002/pssa.202100829. Mater. 34 (2022) 2599–2611, https://doi.org/10.1021/acs.chemmater.1c03809.
[6] M. Nagai, K. Nakanishi, H. Takahashi, H. Kato, T. Makino, S. Yamasaki, [17] Y. Jiang, Z. Deng, B. Zhou, Q. Wei, H. Long, Y. Wang, J. Li, N. Hu, L. Ma, C.-T. Lin,
T. Matsumoto, T. Inokuma, N. Tokuda, Anisotropic diamond etching through Z. Yu, K. Zhou, Nickel-induced transformation of diamond into graphite and
thermochemical reaction between Ni and diamond in high-temperature water carbon nanotubes and the electron field emission properties of resulting composite
vapour, Sci. Rep. 8 (2018) 6687, https://doi.org/10.1038/s41598-018-25193-2. films, Appl. Surf. Sci. 428 (2018) 264–271, https://doi.org/10.1016/j.
[7] W. Smirnov, J.J. Hees, D. Brink, W. Müller-Sebert, A. Kriele, O.A. Williams, C. apsusc.2017.09.129.
E. Nebel, Anisotropic etching of diamond by molten Ni particles, Appl. Phys. Lett. [18] T. Matsumoto, H. Kato, K. Oyama, T. Makino, M. Ogura, D. Takeuchi, T. Inokuma,
97 (2010) 073117, https://doi.org/10.1063/1.3480602. N. Tokuda, S. Yamasaki, Inversion channel diamond metal-oxide-semiconductor
[8] K. Nakanishi, H. Kuroshima, T. Matsumoto, T. Inokuma, N. Tokuda, Atomically flat field-effect transistor with normally off characteristics, Sci. Rep. 6 (2016) 31585,
diamond (100) surface formation by anisotropic etching of solid-solution reaction https://doi.org/10.1038/srep31585.
of carbon into nickel, Diam. Relat. Mater. 68 (2016) 127–130, https://doi.org/ [19] T. Matsumoto, H. Kato, T. Makino, M. Ogura, D. Takeuchi, S. Yamasaki,
10.1016/j.diamond.2016.06.011. T. Inokuma, N. Tokuda, Inversion channel mobility and interface state density of
[9] K. Sakauchi, M. Nagai, T. Tabakoya, Y. Nakamura, S. Yamasaki, C.E. Nebel, diamond MOSFET using N-type body with various phosphorus concentrations,
X. Zhang, T. Matsumoto, T. Inokuma, N. Tokuda, Mechanical damage-free surface Appl. Phys. Lett. 114 (2019) 242101, https://doi.org/10.1063/1.5100328.
planarization of single-crystal diamond based on carbon solid-solution into nickel, [20] T. Matsumoto, T. Yamakawa, H. Kato, T. Makino, M. Ogura, X. Zhang, T. Inokuma,
Diam. Relat. Mater. 116 (2021) 108390, https://doi.org/10.1016/j. S. Yamasaki, N. Tokuda, Fabrication of inversion p-channel MOSFET with a
diamond.2021.108390. nitrogen-doped diamond body, Appl. Phys. Lett. 119 (2021) 242105, https://doi.
[10] T. Tabakoya, M. Nagai, K. Sakauchi, Y. Nakamura, K. Kobayashi, H. Kato, Y. Kato, org/10.1063/5.0075964.
M. Ogura, D. Takeuchi, T. Makino, S. Yamasaki, C.E. Nebel, X. Zhang, [21] Y. Fu, Y. Chang, X. Zhu, R. Xu, H. Kawarada, Normally-off oxidized Si-terminated
T. Matsumoto, T. Inokuma, N. Tokuda, Diamond microfabrication by imprinting (111) diamond MOSFETs via ALD-Al2O3 gate insulator with drain current density
with nickel mold under high temperature, Diam. Relat. Mater. 114 (2021) 108294, over 300 mA/mm, IEEE Trans. On Electron Devices 69 (8) (2022) 4144–4152,
https://doi.org/10.1016/j.diamond.2021.108294. https://doi.org/10.1109/TED.2022.3186865.
[11] Z. Wang, M.R. Shankar, Stability of lithographically patterned transition metal [22] K. Kobayashi, X. Zhang, T. Makino, T. Matsumoto, T. Inokuma, S. Yamasaki, C.-
layers controls spatial selectivity during etching of diamond, Diam. Relat. Mater. E. Nebel, N. Tokuda, Selectively buried growth of heavily B doped diamond layers
130 (2022) 109482, https://doi.org/10.1016/j.diamond.2022.109482. with step-free surfaces in N doped diamond (111) by homoepitaxial lateral growth,
[12] K. Koyama, S.-W. Kim, M. Yoshimoto, Influence of diamond-etching conditions on Appl. Surf. Sci. 593 (2022) 153340, https://doi.org/10.1016/j.
the fabrication of diamond microneedles by a thermochemical reaction of Ni in an apsusc.2022.153340.
H2 atmosphere, JCS-Japan 129 (7) (2021) 481–484, https://doi.org/10.2109/ [23] J. Wang, L. Wan, J. Chen, J. Yan, Surface patterning of synthetic diamond
jcersj2.21025. crystallites using nickel powder, Diam. Relat. Mater. 66 (2016) 206–212, https://
doi.org/10.1016/j.diamond.2016.04.010.