Contents lists available at ScienceDirect

Diamond & Related Materials

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

Ohmic graphite-metal contacts on oxygen-terminated lightly boron-dopedCVD monocrystalline diamond

M. De Feudisa,b,⁎, V. Millea, A. Valentina, O. Brinzaa, A. Tallairea, A. Tardieua, R. Issaouia,J. Acharda,⁎⁎

a Laboratoire des Sciences des Procédés et des Matériaux, CNRS, Université Paris 13, Sorbonne Paris Cité, 93430 Villetaneuse, FrancebDipartimento di Matematica e Fisica, Università del Salento, 73100 Lecce, Italy

A R T I C L E I N F O

Keywords:Ohmic graphite-metal contactsGraphitizationIon implantationOxygen terminationsLightly boron-doped CVD diamond film

A B S T R A C T

A process to obtain ohmic contacts on oxygen-terminated lightly boron-doped CVD monocrystalline diamondfilms was developed. Samples were contacted by Ti/Au metallic pads in the transmission line model (TLM)configuration. The electric contacts were placed onto a mesa structure produced on the CVD boron-doped layer.One of the samples was additionally implanted with helium ions at 10 keV in order to induce the formation of agraphitic layer underneath the diamond surface before contacting so as to improve electrical conduction. Theelectrical performance of both devices was characterized by the TLM method and compared. As a result, thesample with metallic electrodes exhibited a small and non-linear electrical conduction, while the graphitic/metallic contacts showed an ohmic behaviour with an estimated specific contact resistance as low as3.3× 10−4Ω·cm2 for a doping level of a few 1017 cm−3. This approach opens the way to more efficient ohmiccontacts on intrinsic or low-doped diamond that are crucial for the development of electronic devices and de-tectors.

1. Introduction

Nowadays, synthetic CVD diamond films are widely investigated aspower devices [1,2], particle detectors [3,4] and dosimeters [5,6]thanks to their appreciable properties, such as wide band gap, excep-tional thermal conductivity, radiation hardness, high breakdown vol-tage, carrier mobility [1,7]. One of the crucial points for the develop-ment of these diamond devices is the manufacturing of good electriccontacts. In the last decades several studies about the metallic electrodemanufacturing on diamond films, for example with ohmic or Schottkyelectrical behaviour, have been carried out [8–12]. However, up tonow, diamond contacting is still not a simple task due to the inertnessand the wide bang gap of the material. Indeed, diamond contactsshould satisfy different requirements to be considered as good elec-trodes, such as a good mechanical adhesion, chemical, thermal stabilityover time and the possibility of bonding connections. Ohmic contactfabrication, which is the main subject of the present paper, is still anopened field for intrinsic and lightly doped diamonds [13], whilecontacting heavily-doped diamonds is facilitated by the low materialresistivity [12]. Several aspects can affect the performance of metallic

electrodes on diamond films, such as the metallic material used and thediamond surface termination. A typical metallic stacking for diamondohmic contact is composed of titanium and gold layers [2,10,14]. Thetitanium layer ensures titanium carbide (TiC) formation after thermalannealing at about 500 °C [2,14] increasing the contact mechanicaladhesion on the device and the gold layer avoids titanium oxidation. Acrucial role in terms of electrical conduction can also be played by thediamond surface termination. For example, Volpe et al. [8,14] showedthat Ti/Au ohmic contacts were obtained on the H-terminated lightlyboron-doped diamond surface, while by adding an oxidation treatment(by vacuum ultra-violet (VUV) ozone treatment [15]) the electrodesproved to have a Schottky behaviour. But one of the drawback of hy-drogenated surface is its thermal instability and, even if improvementin terms of specific contact resistance can be obtained on such surface,it is highly desirable to develop a deposition process of ohmic contacton lightly boron-doped layer starting from an oxygenated diamondsurface which is well known to be very stable [16].

For low boron-doped oxygen-terminated CVD diamond layers anon-ohmic electrical behaviour is expected, thus a graphitic layer in-duction underneath the diamond surface before the metallic deposition

https://doi.org/10.1016/j.diamond.2018.12.009Received 3 October 2018; Received in revised form 10 December 2018; Accepted 11 December 2018

⁎ Correspondence to: M. De Feudis, Laboratoire des Sciences des Procédés et des Matériaux, CNRS, Université Paris 13, Sorbonne Paris Cité, 93430 Villetaneuse,France.

⁎⁎ Corresponding author.E-mail addresses: mary.defeudis@lspm.cnrs.fr (M. De Feudis), jocelyn.achard@lspm.cnrs.fr (J. Achard).

Diamond & Related Materials 92 (2019) 18–24

Available online 12 December 20180925-9635/ © 2018 Elsevier B.V. All rights reserved.

T

has been considered. Indeed, diamond graphitization just under thesurface can create electrically active defects and consequently reducesthe depletion region width and/or decrease the potential barrier heightat the diamond/metal interface [10]. Diamond graphitization has al-ready been demonstrated by laser or high energy ion implantation onthe diamond surface or into the bulk [7,17,18]. Currently both tech-niques have been successfully used to produce ohmic graphitic contacts[19–25]. A previous study to combine diamond metallization and gra-phitization techniques was carried out by Chen et al. [26] by usingargon ion implantation at 40 keV. They produced ohmic graphite/metalcontacts but with a specific contact resistance which remained rela-tively high (≈ a few Ω.cm−2). In order to improve this result, the mainidea of our work was to produce a graphitic layer just underneath thelightly boron-doped diamond surface using a low ion implantationenergy (10 keV) and a low mass ion source (helium). Subsequently wemanufactured metallic pads in a Transmission Line Model configuration(TLM) [27] on a mesa structure to confine the current lines in the CVDboron-doped active layer. The use of a low boron-doped layer allowsascertaining the charge carrier availability required for the contactelectrical characterization [28].

For completeness, it is worth reporting that in literature, a suc-cessful work about diamond contacting on oxygen-terminated lightlydoped diamond films (2.5× 1017 cm−3) has already been reported.Chen's et al. [29] succeed in ohmic metal contact fabrication using a Ti/Pt/Au deposited on a CVD mesa structure obtaining a very low specificcontact resistance (1.3× 10−5Ω·cm2). Nevertheless nowadays, thir-teen years after this paper, the fabrication of good ohmic metalliccontacts on this kind of diamond film (oxygen-terminated and lightlydoped) is still not a simple task and remains an open field [8,13,14].Therefore, the idea of this work is also to show that, even if the metallicdiamond contacting gives rise to non-ohmic electrical behaviour orohmic one but with a high specific contact resistance, it can be possibleto improve the electrical conduction at the diamond/metal interface byinducing a graphitic layer underneath the diamond surface.

2. Materials and instrumentations

Intrinsic and lightly boron-doped homoepitaxial diamond filmswere grown by Microwave Plasma Assisted Chemical VapourDeposition (MPACVD) technique [30–33]. The reactors were twoBJS150 model provided by Plassys company. One of them is exclusivelydedicated to the growth of intrinsic diamond and has never been con-taminated with boron. In this work intrinsic diamond films were used tostudy the graphitization process while boron doped ones to performelectrical characterizations. The substrates were (100) oriented type-IbHPHT diamonds having dimensions of 3×3×1.5mm3 provided bySumitomo company. The substrate temperature was adjusted to 900 °C,the gas pressure to 140mbar, the microwave power to 2.4 kW and thetotal gas flux to 200 sccm (standard cubic centimeters per min). Themethane and oxygen concentrations were set at 5% and 0.25%, re-spectively, in a purified hydrogen flow. Boron doping has been ensuredby adding diborane (B2H6) as dopant source [34]. Roughly 5 μm-thickfilms were grown. The boron concentration was approximately esti-mated to 4× 1017 cm−3 by cathodoluminescence analyses from thefree to bound exciton ratio in agreement with reference [35]. The CLmeasurements were carried out by using a Scanning Electron Micro-scope (SEM) ZEISS EVO MA-15 coupled with a Horiba-Jobin-YvonHCLUE cathodoluminescence system at a temperature of 110 K. In orderto induce diamond-graphite phase transition underneath the diamondsurface a home-made helium ion implanter was used. This experimentalsetup created a circular ion beam, with a gun diameter of 3mm. Withthis implanter, it is possible to locally modify the pressure in the guneven if the ion flux control remains limited as well as the energy dis-tribution in the irradiated material. For this experiment, the ion ac-celeration energy was set at the low value of 10 keV to limit the pe-netration depth, while the current and the time were set at 300 μA and

30 s, respectively. The implantation dose was estimated to be about1.4×1017 ions·cm−2. This treatment was performed at room tem-perature and in secondary vacuum condition (pressure of 10−5 mbar).Following implantation, a thermal annealing was carried out at 1000 °Cfor 60min in secondary vacuum (10−6 mbar) [36,37]. Metallic contactscomposed of titanium and gold layers in the TLM configuration weredeposited on a CVD diamond mesa structure by several micro-fabrica-tion steps widely discussed in the following Section 3.2. Several char-acterization techniques were used to investigate the samples during allthe processing steps. Structurally investigations were performed by aRaman spectrometer (Jobin-Yvon HR800) using the 632.8 excitationline, morphological analyses were carried out by a SEM (Leica S440)and a profilometer, while diamond surface electrical measurementswere executed by four-point probes method (using a Keithley 4200-SCSmodel). Finally, the metal contacts were electrically studied by the TLMmethod using a Yokogawa 7651 programmable DC source and a FemtoAmp DLCPA-200 amplifier controlled by a LabVIEW software.

3. Results and discussion

3.1. Ion implantation-induced diamond graphitization

The contacting process required to overcome the diamond graphi-tization threshold which is estimated to be about1022–1023 vacancies·cm−3 [38,39]. Based on TRIM (Transport of Ions inMatter [40]) simulations we calculated that under our implantationconditions about 1023 vacancies·cm−3 were produced with a maximumpenetration depth of about 50 nm in diamond. For example, the profileof vacancies generated by an incident helium ion into a diamond film isreported in Fig. 1. The implanted layer seems to spread under thediamond surface at about a depth of 50 nm following a warped Gaus-sian shape distribution.

From an experimental point of view, the ion implantation treatmentwas first tested onto an intrinsic CVD diamond film in order to moreeasily evaluate any electrical and structural change. In Fig. 2, a sampleexposed to the implantation treatment using a ring-like shadow mask tohighlight the optical contrast between the irradiated and unirradiatedzones is shown. The graphitic layer, spatially localized by the mask, isclearly visible since the irradiated zone appears black, in contrast withthe unirradiated one which remained yellow due to nitrogen impuritiesin the HPHT substrate.

To study the structural evolution of the diamond sample exposed tothe ion bombardment, micro-Raman analyses were carried out beforeand after the treatment in particular to estimate, step by step, thediamond-graphite phase change. Micro-Raman spectra registered on

0 20 40 60 80 100 1200.00

0.01

0.02

0.03

0.04

0.05

0.06

Penetration depth (nm)

noirep

ycnacavforeb

muN

(Å-1

)

Fig. 1. TRIM simulation of the vacancy profile produced by a He ion implantedinto a diamond film.

M. De Feudis et al. Diamond & Related Materials 92 (2019) 18–24

19

the initial intrinsic CVD diamond film, after the ion implantation andafter thermal annealing are shown in Fig. 3a, b and c respectively. Atypical Raman spectrum of an intrinsic CVD diamond, with a well-de-fined diamond peak localized at 1332 cm−1 is obtained (Fig. 3a). Onthe contrary, the spectrum of Fig. 3b presents a broad band in the range1300–1600 cm−1, which corresponds to the formation of glassy carbonpeaks containing both amorphous graphite contribution at 1580 cm−1

(G peak) and disordered graphite contribution at 1350 cm−1 (D peak).Then, Fig. 3c shows the structural change of the implanted diamondsurface after thermal annealing. The annealing role is to transform theglassy and amorphous material into graphite for the highly damagedzone, while to restore back the broken diamond bonds in the sur-rounding areas [38]. As expected, the Raman spectrum of the annealedsample shows a modification of the previous 1300–1600 cm−1 bandand the G and D broad peaks sharpen gradually becoming well defined.These peaks confirm the formation of a graphite-like layer [17]. Inaddition, a sharp diamond peak can again be observed due to the re-storing phenomenon.

In order to study the electrical conduction changes induced on in-trinsic CVD monocrystalline diamond by helium implantation, four-point probe current-voltage (I(V)) measurements were carried out.Before doing these measurements, the sample was exposed to a VUV-light irradiation treatment (172 nm) for 45min in an oxygen atmo-sphere at 2mbar in order to avoid any electrical conduction due to theresidual hydrogen termination of as-grown CVD diamond films [15].The first electrical investigation of the intrinsic sample was done beforethe ion implantation, however, the insulator nature of diamond did notallow measuring I(V) curves. I(V) measurements were thus registeredon the sample after both ion implantation and annealing (Fig. 4). Norelevant electrical change was found indicating no specific electricalimprovement of the sp2-bonded graphite as compared with glassy sp2-bonded carbon. The plot shows an electrical conduction with an ohmicbehaviour, which is typical from graphitic materials [7,19,20]. A re-sistance value of 1× 102Ω was extrapolated from the data fitting andconsequently, the sheet resistivity value was estimated equal to 282Ω/□ using Eq. (1) [41]:

=□ρ π F VIln 2

· · (1)

where F is the correction factor which takes into account the samplelimited geometry and the probe position on the surface, which wascomputed equal to 0.6 in a previous work [42] by the Method of Images[43]. This shows that we successfully improved contacts on intrinsicdiamond through ion bombardment-induced graphitization of the sur-face.

For completeness, it must be noted that even if theoretical simula-tions suggests that ion implantation treatment has induced damagesunderneath the diamond surface, more precisely at a depth of about50 nm, it cannot rule out changes in terms of implanted layer thicknessonce the annealing treatment has been carried out. Indeed, in one of our

Fig. 2. Ion implanted CVD monocrystalline intrinsic diamond. A ring-likeshadow mask was used to cover part of the diamond surface.

1000 1200 1400 1600 1800 2000

1000 1200 1400 1600 1800 2000

(a)Intrinsic diamond

Wavenumber (cm-1)

ina

maR

).u.a(ytisnetn

(b)After ion implantation

(c)After annealing

Fig. 3. Raman spectroscopic evolution of the diamond and graphite peaks re-gistered on (a) the as-grown, (b) ion implanted and (c) annealed CVD intrinsicdiamond film.

-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15

-1.0x10-3

-5.0x10-4

0.0

5.0x10-4

1.0x10-3

After ion implantation and annealing Linear fit

)A(tnerru

C

Voltage (V)

Fig. 4. Current-voltage measurements performed with the four-point probemethod and registered on the intrinsic CVD diamond sample after ion im-plantation and annealing. The ohmic behaviour is confirmed by the linear fit ofthe data.

M. De Feudis et al. Diamond & Related Materials 92 (2019) 18–24

20

recent works [42], diamond graphitization process induced by this kindof ion implantation treatment has been carefully studied, finding thatan uncertainty have to be considered between the ion implantationtheoretical forecasts and the final experimental evidence. In particular,while the initial simulations suggested the most damaged layer for-mation at 50 nm underneath the diamond surface, the subsequentanalyses performed by transmission electron microscopy showed thepresence of a 110 nm thick graphitic layer at the diamond surface. Thisdifference was ascribed to the intermedium annealing treatment whichconverted definitely the non-diamond phase into the graphitic one,causing an expansion of the damaged layer due to the different materialdensities. We believe that the same kind of uncertainty has to be takenin consideration for this work.

3.2. Metal contacts and mesa structure manufacturing

Titanium‑gold metal contacts in a TLM configuration with a mesastructure were prepared on two lightly boron-doped diamond films. Themain micro-fabrication technological steps are schematically presentedin Fig. 5.

He+ ion induced graphitization was performed just before anytechnological micro-fabrication steps (Fig. 5, step #1) for one of the 2samples. Then, a silica (SiO2) mask with a thickness of 2 μm and aphotoresist layer were deposited (Fig. 5, steps #2 and #3). An opticalphotolithographic step was performed to transfer the mesa shape ontothe photoresist (Fig. 5, step #4) and afterwards, three plasma etchingsteps using different mixtures of CHF3 and oxygen were performed(Fig. 5, step #5). In particular, the first etching step was carried out onSiO2 layer not protected by the photoresist in order to remove it justaround the mesa structure. Then, the uncovered diamond material wasetched to obtain the mesa structure. In this step we chose for safety toetch for a long time reaching about 13 μm of depth (which is showedsubsequently) even if a shorter treatment would have been sufficientconsidering the CVD layer thickness (5 μm). During this step, the gra-phitic layer around the mesa was removed as the used gas mixture wasreactive with any carbon phase. Finally, an additional SiO2 etching stepwas performed to remove it from the top of the mesa structure.

In Fig. 6 the final CVD diamond mesa profile investigated by aprofilometer is shown. The structure proves to be sharp and 13 μm high.

As the contact mechanical strength on the diamond surface is animportant aspect for both electrical measurement and wire bondingoperations, in order to strengthen this adhesion, a short pre-treatment

of the diamond surface was performed using an argon source before anymetallic deposition steps ([Ar+]= 9 sccm, t=120 s) [2,44] (Fig. 5,step #6). Then, the titanium and gold layers deposition (50 and 200 nmthick, respectively) were carried out (Fig. 5, step #7). By a subsequentphotoresist coating and an optical lithography process the TLM padshape was transferred, while a chemical gold etching (by potassiumiodide solution heated at 50 °C) and a plasma titanium etching (by amixture of Cl2 and Ar) allowed getting the final contacts on CVD mesadiamond (Fig. 5, step #8, #9, #10 and #11, respectively). In particular,these final Au and Ti etching treatments were very important to avoidthe presence of any conductive material among the pads. Before per-forming those on the two diamond samples we optimized the recipes onsilicon wafers after having deposited similar Ti/Au bilayers. It is worthreporting that for these tests the metallic layers were 20% thicker thanones deposited on diamonds (50/200 nm) to ensure the completeetching. Afterwards, an additional diamond etching step was performedby ICP to remove a thin layer of about 150 nm containing the graphiticmaterial among the pads (Fig. 5, step #12). The same recipe was pre-viously tested on the sample in Fig. 2 confirming the complete removalof the graphitic material. Finally, an annealing treatment at 500 °Cduring 60min in secondary vacuum conditions was performed to pro-mote titanium carbide formation (Fig. 5, step #13) [14]. The final paddimensions were 100× 150 μm2 and the space between each pad

Fig. 5. Mesa structure and TLM contact microfabrication steps.

0.0 0.1 0.2 0.3 0.4 0.5-0.002

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

)m

m(thgieH

Width (mm)

Mesa profile

Fig. 6. CVD diamond mesa structure profile registered by profilometry.

M. De Feudis et al. Diamond & Related Materials 92 (2019) 18–24

21

increased from 20 to 50 μm by a step of 10 μm. SEM pictures of a fullprocessed sample are shown in Fig. 7.

3.3. Contact electrical characterization

To characterize electrically both samples, TLM measurements werecarried out and compared. A typical I(V) curve registered on the non-graphitized sample with metallic contacts is presented in Fig. 8. Asexpected for diamond with a low boron doping level, a non-linear be-haviour was clearly obtained. This non-linearity is confirmed for eachpad distance.

On the contrary, as illustrated in Fig. 9a, the sample with graphitic-metallic contact shows an ohmic electrical behaviour whatever the paddistance (d). The electrical resistances were calculated to 540, 700 and920Ω for pad distances of 20, 30 and 40 μm, respectively. The lastcurve (pad distance of 50 μm) and the related electrical resistance valuecould not be investigated due to a micro-fabrication problem causing ashort circuit between the fourth and fifth pads.

The fact that the graphitic layer affects positively the conductivity atthe diamond/metal interface can be ascribed to the generation ofelectrically active defects in agreement with the interface band diagrammodel presented by Tachibana et al. [10]. By this model it has beendemonstrated that the diamond surface treatments producing defects atthe semiconductor/metal interface can reduce the depletion regionwidth increasing the probability of tunneling (Fig. 7b) and/or decreasethe effective potential barrier height promoting the flow of the charge

carriers (Fig. 7c).Subsequently, in order to determine the specific contact resistance,

the total resistance values were extracted from each I(V) curve andplotted as a function of the pad distance d (Fig. 9b). The linear fit fol-lows the Eq. (2) [27]:

= +R R Rw

d2 ·T CSH

(2)

where RT is the total resistance, RC the contact resistance, RSH the sheetresistivity and w the contact width. From the X axis intercept of thislinear fit it was also possible to extrapolate the Transfer Length LTwhich corresponds to the real contact length crossed by the currentlines. It is worth noting that for contact length L≥ 1.5 LT, which is thiscase, the specific contact resistance ρC can be estimated by the Eq. (3)[41]:

=ρ R L w· ·C C T (3)

Definitely, considering that RC and LT values proved to be equal to66Ω and 3.3 μm, respectively, the estimated specific contact resistanceρC was about 3.3× 10−4Ω·cm2, namely one of the lowest values re-ported for oxygen-terminated lightly doped diamonds [13,26]. Thisdifferent electrical behaviour between the two samples can be ascribedto the generation of electrically active graphitic defects in the ion im-planted diamond which improve the charge carrier flow at the semi-conductor/metal interface. Therefore, this diamond contacting methodcould open the way to the fabrication of ohmic graphitic-metalliccontacts on oxygen-terminated and lightly doped diamonds for dif-ferent applications such as the n-type doped films which are still anopened field.

4. Conclusion

We used helium ion implantation treatment (10 keV,1.4× 1017 ions·cm−2) followed by annealing to generate a graphitelayer underneath the surface of different intrinsic or lightly boron-doped diamond single crystal CVD films. Raman spectroscopic analysesconfirmed the formation of sp2 bonded carbon and 4-point probemeasurements led to a sheet resistivity of 282Ω/□ for an intrinsic film.In order to assess the benefit of using such shallow implanted graphiticlayers on electrical contacts, lightly boron-doped oxygen-terminateddiamond samples were contacted by Ti/Au contacts in a TLM config-uration over a mesa structure. Current-voltage characterizations of themetal contacts have been performed showing that, without the gra-phitization step, non-linear behaviour was registered. On the contrary,with the additional He+ implantation step, a linear behaviour wasobtained leading to a specific contact resistance of 3.3× 10−4Ω·cm2.

(a)

1 mm

(b)

100 µm

Fig. 7. CVD boron-doped diamond with TLM-like metallic contacts. (a) Lightly tilted top view of the full sample: the central mesa/TLM zone surrounding by differentadditional patterns is visible. (b) Tilted enlargement of the mesa/TLM zone (in the center of the picture) showing the height of the area with respect to thesurrounding surface.

-1.0 -0.5 0.0 0.5 1.0

-6.0x10-7

-4.0x10-7

-2.0x10-7

0.0

2.0x10-7

4.0x10-7

6.0x10-7

8.0x10-7

)A(tnerru

C

Voltage (V)

Fig. 8. A typical current-voltage curve for the lightly boron-doped sample withmetal contacts (pad distance of 20 μm).

M. De Feudis et al. Diamond & Related Materials 92 (2019) 18–24

22

This improvement can be ascribed to the generation of electrically ac-tive defects (graphite) at the diamond/metal interface which can re-duce the depletion region width increasing the probability of tunnelingand/or decrease the effective potential barrier height promoting theflow of the charge carriers. Therefore, even if this specific contact re-sistance value remains rather high for electronic devices, it can beconsidered as a good starting point taking into account the low borondoping level and the oxygen-terminated surface of diamond. Definitely,it is expected that this method of diamond contacting could be extendedto n-type diamond and thus could open the way to the production ofohmic graphitic - metallic contacts for different applications.

Acknowledgement

We acknowledge Dr. Philippe Lafarge, Dr. Maria Luisa Della Roccaand Dr. Clément Barraud (TELEM group, University of Paris Diderot)for their valuable collaboration about the electrical measurements andthe clean room team of University Paris Diderot for the Ti, Au, SiO2 filmdepositions on our samples.

This work has been supported by Region Ile-de-France in the fra-mework of DIM SIRTEQ.

References

[1] R.S. Balmer, J.R. Brandon, S.L. Clewes, H.K. Dhillon, J.M. Dodson, I. Friel,P.N. Inglis, T.D. Madgwick, M.L. Markham, T.P. Mollart, N. Perkins,G.A. Scarsbrook, D.J. Twitchen, A.J. Whitehead, J.J. Wilman, S.M. Woollard,Chemical vapour deposition synthetic diamond: materials, technology and appli-cations, J. Phys. Condens. Matter 21 (36) (2009) 364221.

[2] L.S. Pan, D.R. Kania, e.H.L.T (Ed.), Diamond: Electronic Properties andApplications, Springer science+Business Media LLC ed., New York, USA, 1995.

[3] S.N. Ahmed, Physics and Engineering of Radiation Detection, Academic Press is animprint of Elsevier ed., Great Britain, 2007.

[4] M. Alison, Recent developments of diamond detectors for particles and UV radia-tion, Semicond. Sci. Technol. 15 (9) (2000) R55.

[5] A. Bartoli, I. Cupparo, A. Baldi, M. Scaringella, A. Pasquini, S. Pallotta,C. Talamonti, M. Bruzzi, Dosimetric characterization of a 2D polycrystalline CVDdiamond detector, J. Instrum. 12 (03) (2017) C03052.

[6] S. Almaviva, M. Marinelli, E. Milani, A. Tucciarone, G. Verona-Rinati, R. Consorti,A. Petrucci, F. De Notaristefani, I. Ciancaglioni, Synthetic single crystal diamonddiodes for radiotherapy dosimetry, Nucl. Instrum. Methods Phys. Res., Sect. A 594(2) (2008) 273–277.

[7] V.I. Konov, Laser in micro and nanoprocessing of diamond materials, LaserPhotonics Rev. 6 (6) (2012) 739–766.

[8] P.-N. Volpe, P. Muret, J. Pernot, F. Omnès, T. Teraji, Y. Koide, F. Jomard,D. Planson, P. Brosselard, N. Dheilly, B. Vergne, S. Scharnholz, Extreme dielectricstrength in boron doped homoepitaxial diamond, Appl. Phys. Lett. 97 (22) (2010)223501.

[9] H. Umezawa, K. Yukako, S. Shin-ichi, 1Ω on-resistance diamond vertical-Schottkybarrier diode operated at 250 °C, Appl. Phys. Express 6 (1) (2013) 011302.

[10] T. Tachibana, B.E. Williams, J.T. Glass, Correlation of the electrical properties ofmetal contacts on diamond films with the chemical nature of the metal-diamondinterface. II. Titanium contacts: a carbide-forming metal, Phys. Rev. B 45 (20)(1992) 11975–11981.

[11] T. Tachibana, B.E. Williams, J.T. Glass, Correlation of the electrical properties ofmetal contacts on diamond films with the chemical nature of the metal-diamondinterface. I. Gold contacts: a non-carbide-forming metal, Phys. Rev. B 45 (20)(1992) 11968–11974.

[12] T. Makino, O. Kazuhiro, K. Hiromitsu, T. Daisuke, O. Masahiko, O. Hideyo,

Y. Satoshi, Diamond electronic devices fabricated using heavily doped hopping p+and n+ layers, Jpn. J. Appl. Phys. 53 (2014) 05FA12.

[13] T. Teraji, S. Koizumi, Y. Koide, Ohmic contact for p-type diamond without post-annealing, J. Appl. Phys. 104 (1) (2008) 016104.

[14] P.N. Volpe, P. Muret, J. Pernot, F. Omnès, T. Teraji, F. Jomard, D. Planson,P. Brosselard, N. Dheilly, B. Vergne, S. Scharnholz, High breakdown voltageSchottky diodes synthesized on p-type CVD diamond layer, Phys. Status Solidi A207 (9) (2010) 2088–2092.

[15] T. Teraji, Y. Garino, Y. Koide, T. Ito, Low-leakage p-type diamond Schottky diodesprepared using vacuum ultraviolet light/ozone treatment, J. Appl. Phys. 105 (12)(2009) 126109.

[16] B.B. Pate, M.H. Hecht, C. Binns, I. Lindau, W.E. Spicer, Photoemission and photon-stimulated ion desorption studies of diamond (111): hydrogen, J. Vac. Sci. Technol.21 (1982) 364.

[17] R. Kalish, A. Reznik, S. Prawer, D. Saada, J. Adler, Ion-implantation-induced defectsin diamond and their annealing: experiment and simulation, Phys. Status Solidi A174 (83) (1999) 83–99.

[18] A. Taurino, M. Catalano, M. De Feudis, A.P. Caricato, M. Martino, Q. Wang,M.J. Kim, Graphitization of diamond by means of UV laser writing: a transmissionElectron microscopy study, Microsc. Microanal. 23 (S1) (2017) 2262–2263.

[19] M. De Feudis, A.P. Caricato, G. Chiodini, M. Martino, E. Alemanno, G. Maruccio,A.G. Monteduro, P.M. Ossi, R. Perrino, S. Spagnolo, Characterization of surfacegraphitic electrodes made by excimer laser on CVD diamond, Diam. Relat. Mater.65 (2016) 137–143.

[20] M. De Feudis, A.P. Caricato, A. Taurino, P.M. Ossi, C. Castiglioni, L. Brambilla,G. Maruccio, A.G. Monteduro, E. Broitman, G. Chiodini, M. Martino, Diamondgraphitization by laser-writing for all-carbon detector applications, Diam. Relat.Mater. 75 (2017) 25–33.

[21] S. Salvatori, P. Oliva, M. Pacilli, P. Allegrini, G. Conte, M. Komlenok, A.A. Khomich,A. Bolshakov, V. Ralchenko, V. Konov, Nano-carbon pixels array for ionizing par-ticles monitoring, Diam. Relat. Mater. 73 (2017) 132–136.

[22] M. Komlenok, A. Bolshakov, V. Ralchenko, V. Konov, G. Conte, M. Girolami,P. Oliva, S. Salvatori, Diamond detectors with laser induced surface graphite elec-trodes, Nucl. Instrum. Methods Phys. Res., Sect. A 837 (2016) 136–142.

[23] J.C. Piñero, M.P. Villar, D. Araujo, J. Montserrat, B. Antúnez, P. Godignon, Impactof thermal treatments in crystalline reconstruction and electrical properties ofdiamond Ohmic contacts created by boron ion implantation, Phys. Status Solidi A214 (11) (2017) 1700230.

[24] A.P. Bolshakov, K.N. Zyablyuk, V.A. Kolyubin, V.A. Dravin, R.A. Khmelnitskii,P.G. Nedosekin, V.N. Pashentsev, E.M. Tyurin, V.G. Ralchenko, Thin CVD diamondfilm detector for slow neutrons with buried graphitic electrode, Nucl. Instrum.Methods Phys. Res., Sect. A 871 (2017) 142–147.

[25] T. Kononenko, V. Ralchenko, A. Bolshakov, V. Konov, P. Allegrini, M. Pacilli,G. Conte, E. Spiriti, All-carbon detector with buried graphite pillars in CVD dia-mond, Appl. Phys. A 114 (2) (2014) 297–300.

[26] Y.G. Chen, M. Hasegawa, H. Okushi, S. Koizumi, H. Yoshida, T. Sakai, N. Kobayashi,Electrical properties of graphite/homoepitaxial diamond contact, Diam. Relat.Mater. 11 (3) (2002) 451–457.

[27] G.K. Reeves, H.B. Harrison, Obtaining the specific contact resistance from trans-mission line model measurements, IEEE Electron Device Lett. 3 (5) (1982) 111–113.

[28] V. Mortet, J. Pernot, F. Jomard, A. Soltani, Z. Remes, J. Barjon, J. D'Haen,K. Haenen, Properties of boron-doped epitaxial diamond layers grown on (110)oriented single crystal substrates, Diam. Relat. Mater. 53 (2015) 29–34.

[29] Y. Chen, M. Ogura, S. Yamasaki, H. Okushi, Ohmic contacts on p-type homo-epitaxial diamond and their thermal stability, Semicond. Sci. Technol. 20 (2005)860–863.

[30] M. Kamo, Y. Sato, S. Matsumoto, N. Setaka, Diamond synthesis from gas phase inmicrowave plasma, J. Cryst. Growth 62 (3) (1983) 642–644.

[31] J. Achard, F. Silva, A. Tallaire, X. Bonnin, G. Lombardi, K. Hassouni, A. Gicquel,High quality MPACVD diamond single crystal growth: high microwave powerdensity regime, J. Phys. D. Appl. Phys. 40 (20) (2007) 6175.

[32] A. Tallaire, V. Mille, O. Brinza, T.N. Tran Thi, J.M. Brom, Y. Loguinov, A. Katrusha,A. Koliadin, J. Achard, Thick CVD diamond films grown on high-quality type IIaHPHT diamond substrates from new diamond technology, Diam. Relat. Mater. 77(2017) 146–152.

[33] J. Achard, A. Tallaire, Growth of thick CVD diamond films on different crystalline

-0.10 -0.05 0.00 0.05 0.10-2.0x10-4

-1.5x10-4

-1.0x10-4

-5.0x10-5

0.05.0x10-5

1.0x10-4

1.5x10-4

2.0x10-4 d = 20 md = 30 md = 40 m)

A(tnerruC

Voltage (V)

(a)

0 5 10 15 20 25 30 35 40 450

1x1022x1023x1024x1025x1026x1027x1028x1029x1021x1031x103(b)

ResistanceLinear fit

)mh

O(ecnatsiserlato

T

Pad distance ( m)µ

µµµ

Fig. 9. (a) Current-voltage curves for the lightly boron-doped sample with graphite-metal contacts for different pad distances d (blue squares 20 μm, red circles 30 μmand green triangles 40 μm). (b) Total resistance as a function of the pad distance and linear fit.

M. De Feudis et al. Diamond & Related Materials 92 (2019) 18–24

23

orientations: defects and doping, in: S. Koizumi, et al. (Ed.), Power ElectronicsDevice Applications of Diamond Semiconductors, Elsevier Ltd., United Kingdom,2018.

[34] R. Issaoui, J. Achard, F. Silva, A. Tallaire, V. Mille, A. Gicquel, Influence of oxygenaddition on the crystal shape of CVD boron doped diamond, Phys. Status Solidi A208 (9) (2011) 2023–2027.

[35] J. Barjon, T. Tillocher, N. Habka, O. Brinza, J. Achard, R. Issaoui, F. Silva, C. Mer,P. Bergonzo, Boron acceptor concentration in diamond from excitonic recombina-tion intensities, Phys. Rev. B 83 (7) (2011) 073201.

[36] F. Bosia, N. Argiolas, M. Bazzan, P. Olivero, F. Picollo, A. Sordini, M. Vannoni,E. Vittone, Modification of the structure of diamond with MeV ion implantation,Diam. Relat. Mater. 20 (5) (2011) 774–778.

[37] V.S. Drumm, A.D.C. Alves, B.A. Fairchild, K. Ganesan, J.C. McCallum,D.N. Jamieson, S. Prawer, S. Rubanov, R. Kalish, L.C. Feldman, Surface damage ondiamond membranes fabricated by ion implantation and lift-off, Appl. Phys. Lett. 98(23) (2011) 231904.

[38] C. Uzan-Saguy, C. Cytermann, R. Brener, V. Richter, M. Shaanan, R. Kalish, Damagethreshold for ion-beam induced graphitization of diamond, Appl. Phys. Lett. 67 (9)

(1995) 1194–1196.[39] P. Olivero, S. Rubanov, P. Reichart, B.C. Gibson, S.T. Huntington, J.R. Rabeau,

A.D. Greentree, J. Salzman, D. Moore, D.N. Jamieson, S. Prawer, Characterizationof three-dimensional microstructures in single-crystal diamond, Diam. Relat. Mater.15 (10) (2006) 1614–1621.

[40] J.F. Ziegler, M. Ziegler, J. Biersack, SRIM – the stopping and range of ions in matter(2010), Nucl. Inst. Methods Phys. Res. B 268 (11−12) (2010) 1818–1823.

[41] D.K. Schroder, Semiconductor Material and Device Characterization, Wiley-IEEEPress ed, Hoboken, New Jersey, 2006.

[42] A. Valentin, M. De Feudis, O. Brinza, A. Tardieu, L. William, A. Tallaire, J. Achard,Characteristics of He ion implanted layers on single-crystal diamond, Phys. StatusSolidi A 215 (2018) 1800264.

[43] L.B. Valdes, Resistivity measurements on germanium for transistors, Proc. IRE 42(2) (1954) 420–427.

[44] T. Tachibana, J.T. Glass, Effects of argon presputtering on the formation of alu-minum contacts on polycrystalline diamond, J. Appl. Phys. 72 (12) (1992)5912–5918.

M. De Feudis et al. Diamond & Related Materials 92 (2019) 18–24

24