- Email: [email protected]
Reprint of "Palladium Ohmic contact on hydrogen-terminated single crystal diamond film"
Diamond & Related Materials 63 (2016) 175–179
Contents lists available at ScienceDirect
Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond
Reprint of "Palladium Ohmic contact on hydrogen-terminated single crystal diamond film"☆ W. Wang, C. Hu, F.N. Li, S.Y. Li, Z.C. Liu, F. Wang, J. Fu, H.X. Wang ⁎ Key Laboratory for Physical Electronics and Devices of the Ministry of Education, The School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, China
a r t i c l e
i n f o
Article history: Received 15 July 2015 Received in revised form 21 September 2015 Accepted 21 September 2015 Available online 5 February 2016 Keywords: Palladium Ohmic contact Specific contact resistance XPS Annealing Surface treatment
a b s t r a c t Contact properties of Palladium (Pd) on the surface of hydrogen-terminated single crystal diamond were investigated with several treatment conditions. 150 nm Pd pad was deposited on diamond surface by thermal evaporation technique, which shows good Ohmic properties with the specific contact resistivity (ρc) of 1.8 × 10−6 Ω cm2 evaluated by Transmission Line Model. To identify the thermal stability, the sample was annealed in Ar ambient from 300 to 700 °C for 3 min at each temperature. As the temperature increased, ρc firstly decreased to 4.93 × 10−7 Ω cm2 at 400 °C and then increased. The barrier height was evaluated to be −0.15 eV and − 0.03 eV for as-deposited and 700 °C annealed sample by X-ray photoelectron spectroscopy analysis. Several surface treatments were also carried out to determine their effect on ρc, among which HNO3 vapor treated sample indicates a lower value of 5.32 × 10−6 Ω cm2. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Diamond appears promising for high-power and high-frequency devices, since it has remarkable properties, such as wide band gap (5.47 eV), highest thermal conductivity (22 W/cm K), high breakdown field (N 10 MV/cm), high carrier mobility of electron (4500 cm2 V− 1 s− 1) and hole (3800 cm2 V− 1 s− 1), and high saturation velocity (10 7 cm/s) [1–2]. Recently, applicability of semiconducting diamond films to electronic devices has been extensively studied by fabricating Schottky diodes [3–7] and field-effect transistors (FET) [8–15]. One of the most essential topics to be studied for the diamond applications is the contact properties between the electrode metal layer and the diamond film. Up to now, many refractory metals such as Ti, Mo and Ta, which are capped by Au [12,16,17], are used as contact material for p-type diamond, on account of carbide formation at the interface with the annealing temperature higher than 400 °C. Au [18], a non-carbide-forming metal, was also used as contact metal, whereas the adherence of Au on diamond film is quite poor and could peel off easily even during lift-off process without ultrasonic. However, few investigations on Pd Ohmic contact to hydrogen-terminated single crystal diamond film were reported. ☆ A Publisher’s error resulted in this article appearing in the wrong issue. The article is reprinted here for the continuity of the Special Issue: 9th International Conference on New Diamond and New Carbons (NDNC) 2015. For citation purposes, please use the original publication details; Diamond and Related Materials, Volume 59, pp. 90–94. DOI of original article: http://dx.doi.org/10.1016/j.diamond.2015.09.012. ⁎ Corresponding author. E-mail address: [email protected] (H.X. Wang).
http://dx.doi.org/10.1016/j.diamond.2016.01.019 0925-9635/© 2015 Elsevier B.V. All rights reserved.
In this study, contact properties of Pd on hydrogen terminated diamond surface were investigated. The specific contact resistivity (ρc) was evaluated by Transmission Line Model (TLM). Then, annealing process was performed to examine the thermal stability of the contact. Furthermore, the X-ray photoelectron spectroscopy (XPS) technique was used to identify the contact barrier height at the interface. For comparison, before Pd TLM patterns deposition, the other diamond samples were treated by several kinds of solutions and vapors. 2. Experiment Undoped homoepitaxial diamond films were grown by microwave plasma assisted CVD onto high pressure and high-temperature synthetic Ib diamond (100) substrates with the dimension of 3 × 3 × 0.5 mm3. The total flow rate of the reaction gas was 500 sccm, and the ratio of CH4/H2 was 1%. The process pressure, growth temperature and microwave power were 100 Torr, 900 °C and 1 kW, respectively. After growth, hydrogen plasma was kept for 10 min to form the hydrogen termination, and the samples were cooled down in pure hydrogen ambient. The Hall measurement, Raman spectra, X-Ray Diffraction (XRD) and Atomic Force Microscope (AFM) were carried out to evaluate the epitaxial diamond film. After 3 nm Pd being evaporated on diamond surface, the fabrication process was implemented in two different ways, which is shown in Fig. 1. In one way, 150 nm Pd TLM configuration with an area of 100 × 100 μm2 and space ranging from 5 to 30 μm was firstly formed on diamond film by using the lithography and thermal evaporation method. A negative photoresist (PR) of Az5214 was used, whose prebake and reversal bake temperature were 95 °C and
176
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
Fig. 1. The fabrication process flow of the TLM configuration, a) sample A for annealing process, and b) sample B for surface treatment.
110 °C, respectively. Secondly, the sample was treated by UV/ozone for isolation. Thirdly, the sample was annealed to investigate its thermal stability in Ar (5 N) ambient at several temperatures from 300 to 700 °C during which Ar flow rate was set to 2 L/min and the pressure was kept at atmospheric pressure. Finally, XPS was utilized to determine the barrier height of the Pd contact on hydrogenterminated diamond. This sample was denoted as sample A. In another way, samples were firstly treated by different kinds of solutions and vapors after lithography, including HCl, NH3·H2O and HNO3. Secondly,
150 nm Pd TLM pads were evaporated onto the diamond film. Thirdly, samples were treated by UV/ozone. These samples were named as sample B group. 3. Results and discussion The sample morphologies were investigated by optical microscope (OM) and AFM. The typical images are shown in Fig. 2. Fig. 2(a) shows the full-scale OM image, illustrating a smooth
Fig. 2. a) The optical microscope image of as-deposited single crystal diamond film with the dimension of 3 × 3 mm2, b) AFM image with the area of 5 × 5 μm2, c) the optical microscope image of TLM configuration after annealing at 700 °C in air for 3 min.
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
surface. Fig. 2(b) indicates AFM measurement result, exhibiting a very small surface roughness of 0.318 nm. The mobility, square resistance, carrier density determined by Hall measurement are 92 cm 2 V− 1 s − 1 , 6.5 kΩ, 1.04 × 1013 cm − 2 , respectively. The XRD rocking curve of (400) crystal plane and Raman spectra indicate the good crystalline quality and pure sp 3 structure of the film with relatively low FWHM of 0.01 ° and 2.859 cm− 1, respectively. Fig. 2(c) shows the OM image of sample A after being annealed at 700 °C in Ar atmosphere for 3 min. The edges of all TLM pads are neat, and there is no peeling off phenomenon, indicating good thermal stability. Also, the Pd pads were subjected to ultrasonic during the lift-off process, exhibiting pretty well adherence. In this image, there are some dark features that appeared on the TLM pads, which is ascribed to the scratches induced by the probe. To investigate the thermal stability of the Pd contact on diamond, sample A was annealed by rapid thermal process at several temperatures from 300 to 700 °C in Ar for 3 min. Fig. 3 shows the variation of ρc with the annealing temperature, which is an average value of 6 TLM configurations. As the temperature increases, ρc starts to decrease and reaches the minimum of 4.93 × 10 − 7 Ω cm2 at 400 °C. Compared with ρc of as-deposited sample, the minimum value obtained at 400 °C reduces no more than one order of magnitude, indicating good thermal stability of the Ohmic contact. The initial reduction of ρc may be related with the modification of Pd and diamond interface after annealing. With further increase of the temperature, ρc begins to increase and reaches the maximum at 700 °C with the value of 9.68 × 10− 6 Ω cm 2 , which is larger than that of as-deposited sample. And it's worth noting that the Pd contact on H-terminated diamond can perform excellent Ohmic properties even without annealing. As displayed in Fig. 4(a), the linear fitting curves of the sample annealed at 400 °C and as-deposited sample exhibit a high degree of similarity. The I–V curves in Fig. 4(b) reveal little difference between sample annealed at 700 °C and the as-deposited sample, illustrating no obvious degradation of the Ohmic behavior after such high temperature. Meanwhile, the surface conduction is not reduced, which indicates that the hydrogen termination has been well preserved under such harsh condition. The XPS measurement was used to determine the barrier height of the Pd contact on the diamond film. As shown in Fig. 5, for asdeposited sample, the X-ray beam first focused on the thin Pd to investigate C 1 s and Pd 3d peaks, then X-ray beam was focused on the thick Pd to measure Pd 3d peaks so as to calibrate the C 1s peak of thin Pd. For annealed sample, the X-ray beam first focused on the edge of the Pd pad to investigate C 1s and Pd 3d peaks, then X-ray beam was focused on the Pad to measure Pd 3d peaks so as to calibrate the C 1s peak of Edge. The XPS spectrum of C 1s and Pd 3d are displayed in Fig. 6. Compared with thick Pd and Pad 3d peaks, the 3d peaks of thin Pd and Edge indicate
Fig. 3. The temperature dependence of the specific contact resistance. The annealing temperature is from 300 °C to 700 °C with the atmosphere of Argon for 3 min at each temperature.
177
Fig. 4. a) Linear fit of the same TLM pattern before and after annealing in Ar at 300 °C for 3 min, b) I–V plots of the same two Pd pads before and after annealing in Ar at 700 °C for 3 min.
blue-shift of 0.22 eV and 0.25 eV, respectively. Thus, based on the peak values in Fig. 6(b), the core-level of C 1s is calibrated to be 283.97 − 0.22 = 283.75 eV and 284.13 − 0.25 = 283.88 eV for thin Pd and Edge, respectively. As a result, the barrier height of asdeposited sample is estimated to be −0.15 ± 0.1 eV by using the equation ΦBH = EC-1s-XPS − EC-1s-Rf [19]. The reference binding energy of C 1s (EC-1s-Rf) was determined to be 283.9 ± 0.1 eV with respect to EF [20]. The band diagram of the contact is displayed in Fig. 7. As the barrier height is −0.15 eV, the valance band maximum at the interface bends up. So the holes can easily flow to metal without any barrier, which leads to the excellent Ohmic behavior of Pd contact on the diamond film. Compared with that of as-deposited sample in Fig. 6(a), the Pd 3d peak of annealed sample is scarcely shifted, indicating that there was no carbide formed at the interface of contact. It is one of the reasons that annealing has a little effect on the specific contact resistivity. With the same method mentioned above, the barrier height of 700 °C annealed sample is calculated to be − 0.03 ± 0.1 eV with C 1s peak value of 284.13 eV. The increase of the barrier height gives an explanation why the ρc at 700 °C becomes higher than that of as-deposited. The ρc of sample B group under different surface treatment is revealed in Table 1. Only the treatment of HNO3 vapor can reduce the value of ρc, which could be ascribed to the existence of NO2 in HNO3 atmosphere [21]. However, the other solutions and vapor lead to the degradation of Ohmic properties. One of the possible reason is that
178
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
Fig. 5. The schematic diagram of XPS testing for evaluation of barrier height.
when diamond surface adsorbed HCl and NH3·H2O, the electron is not easily transferred from diamond valance band to adsorbate layer, resulting in low density of 2 dimensional hole gas and lower surface conduction.
4. Conclusion Contact properties of Palladium (Pd) on the surface of hydrogenterminated single crystal diamond have been investigated with several treatment conditions. The ρc of as-deposited sample has been evaluated to be 1.8 × 10−6 Ω cm2. As sample A being annealed from 300 to 700 °C, ρc firstly decreased and then increased. The I–V curves for as-deposited
and annealed sample indicate little difference, illustrating no degradation of Ohmic properties and preserved hydrogen termination. The barrier height of the contact was identified to be − 0.15 ± 0.1 eV and −0.03 ± 0.1 eV for as-deposited and annealed sample by the method of XPS spectra, respectively. And no carbide formation was observed, showing good thermal stability of the contact. Different surface treatments were applied to the sample B and HNO3 vapor was found to enhance the Ohmic properties, the reason may be the existence of NO2 vapor. Prime novelty statement This work further investigates the Pd Ohmic contact properties on intrinsic diamond surface, getting a relatively low specific contact resistance (ρc) of 1.8 × 10−6 Ω cm2. Annealing process with different temperatures indicates the thermal stability of the contact. It's worth noting that the hydrogen termination is preserved after 700 °C annealing. The XPS analysis was carried out to define the barrier heights of as-deposited and annealed samples, explaining the increase of ρc after 700 °C annealing. Also, the surface treatments were applied to verify the effect on ρc.
Fig. 7. The band diagram at the interface of Pd and diamond.
Table 1 The specific Ohmic contact resistance of Pd on single crystal diamond film with different surface treatment. ρc (Ω cm2)
Fig. 6. The XPS spectrums of a) Pd 3d peaks and b) C 1s peaks for as-deposited and annealed samples.
As-deposited HCl NH3·H2O HNO3(vapor) HCl (vapor)
2.14 × 10−5 2.21 × 10−3 2.00 × 10−4 5.35 × 10−6 1.09 × 10−3
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
Acknowledgments This paper was supported by the National High-tech R&D Program of China (863 Program) (Grant No.2013AA03A101). References [1] J. Isberg, J. Hammersberg, E. Johansson, T. Wikstrom, D.J. Twitchen, A.J. Whitehead, S.E. Coe, A. Scarsbrook, High carrier mobility in single-crystal plasma-deposited diamond, Science 297 (2002) 1670–1672. [2] L. Reggiani, S. Bosi, C. Canai, F. Nava, Hole-drift velocity in natural diamond, Phys. Rev. B Condens. Matter 23 (1981) 3050–3057. [3] H. Umezawa, S. Shikata, T. Funaki, Diamond Schottky barrier diode for hightemperature, high-power, and fast switching applications, Jpn. J. Appl. Phys. 53 (2014) (05FP06). [4] A. Vescan, I. Daumiller, P. Gluche, W. Ebert, E. Kohn, High temperature, high voltage operation of diamond Schottky diode, Diam. Relat. Mater. 7 (1998) 581–584. [5] D. Takeuchi, S. Yamanaka, H. Okushi, Schottky junction properties of the high conductivity layer of diamond, Diam. Relat. Mater. 11 (2002) 355–358. [6] H. Umezawa, N. Tokuda, M. Ogura, S. Ri, S. Shikata, Characterization of leakage current on diamond Schottky barrier diodes using thermionic-field emission modeling, Diam. Relat. Mater. 15 (2006) 1945–1953. [7] A. Traore, P. Muret, A. Fiori, D. Eon, E. Gheeraert, J. Pernot, Zr/oxidized diamond interface for high power Schottky diodes, Appl. Phys. Lett. 104 (2014) 052105. [8] H. Kawarada, M. Aoki, H. Sasaki, K. Tsugawa, Characterization of hydrogenterminated CVD diamond surfaces and their contact properties, Diam. Relat. Mater. 3 (1994) 961–965. [9] K. Tsugawa, K. Kitatani, H. Noda, A. Hokazono, K. Hirose, M. Tajima, H. Kawarada, High-performance diamond surface-channel field-effect transistors and their operation mechanism, Diam. Relat. Mater. 8 (1999) 927–933. [10] H. Kawarada, High-current metal oxide semiconductor field-effect transistors on H-terminated diamond surfaces and their high-frequency operation, Jpn. J. Appl. Phys. 51 (2012) 090111.
179
[11] A. Daicho, T. Saito, S. Kurihara, A. Hiraiwa, H. Kawarada, High-reliability Passivation of Hydrogen-terminated Diamond Surface by Atomic Layer Deposition of Al2O3, 115 2014, p. 223711. [12] Y. Jingu, K. Hirama, H. Kawarada, Ultrashallow TiC source/drain contacts in diamond MOSFETs formed by hydrogenation-last approach, IEEE Trans. Electron Devices 57 (2010) 966–972. [13] K. Ueda, M. Kasu, Y. Yamauchi, T. Makimoto, M. Schwitters, D.J. Twitchen, G.A. Scarsbrook, S.E. Coe, Diamond FET using high-quality polycrystalline diamond with fT of 45 GHz and fmax of 120 GHz, IEEE Electron Device Lett. 27 (2006) 570–572. [14] K. Hirama, H. Sato, Y. Harada, H. Yamamoto, M. Kasu, Diamond field-effect transistors with 1.3 A/mm drain current density by Al2O3 passivation layer, Jpn. J. Appl. Phys. 51 (2012) 090112. [15] A. Stephen, O. Russell, S. Sharabi, A. Tallaire, D.A.J. Moran, Hydrogen-terminated diamond field-effect transistors with cutoff frequency of 53 GHz, IEEE Electron Device Lett. 33 (2012) 1471–1473. [16] C.M. Zhen, Y.Y. Wang, S.H. He, Q.F. Guo, Z.J. Yan, Y.J. Pu, Ohmic contacts to borondoped diamond, Opt. Mater. 23 (2003) 117–121. [17] J. Nakanishi, A. Otsuki, T. Oku, O. Ishiwata, M. Murakami, Formation of ohmic contacts to P-type diamond using carbide forming metals, J. Appl. Phys. 76 (1994) 2293–2298. [18] C.M. Zhen, X.Q. Wang, X.C. Wu, C.X. Liu, D.L. Hou, Au/p-diamond ohmic contacts deposited by RF sputtering, Appl. Surf. Sci. 255 (2008) 2916–2919. [19] S. Kono, T. Nohara, S. Abe, H. Kodama, K. Suzuki, S. Koizumi, T. Abukawa, A. Sawabe, Electron spectroscopic determination of electronic structures of phosphorus-doped n-type heteroepitaxial diamond (001) surface and junction, Jpn. J. Appl. Phys 51 (2012) 090109. [20] F. Maier, J. Ristein, L. Ley, Electron affinity of plasma-hydrogenated and chemically oxidized diamond (100) surfaces, Phys. Rev. B 64 (2001) 165411. [21] M. Kubovic, M. Kasu, H. Kageshima, Sorption properties of NO2 gas and its strong influence on hole concentration of H-terminated diamond surfaces, Appl. Phys. Lett. 96 (2010) 052101.
Contents lists available at ScienceDirect
Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond
Reprint of "Palladium Ohmic contact on hydrogen-terminated single crystal diamond film"☆ W. Wang, C. Hu, F.N. Li, S.Y. Li, Z.C. Liu, F. Wang, J. Fu, H.X. Wang ⁎ Key Laboratory for Physical Electronics and Devices of the Ministry of Education, The School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, China
a r t i c l e
i n f o
Article history: Received 15 July 2015 Received in revised form 21 September 2015 Accepted 21 September 2015 Available online 5 February 2016 Keywords: Palladium Ohmic contact Specific contact resistance XPS Annealing Surface treatment
a b s t r a c t Contact properties of Palladium (Pd) on the surface of hydrogen-terminated single crystal diamond were investigated with several treatment conditions. 150 nm Pd pad was deposited on diamond surface by thermal evaporation technique, which shows good Ohmic properties with the specific contact resistivity (ρc) of 1.8 × 10−6 Ω cm2 evaluated by Transmission Line Model. To identify the thermal stability, the sample was annealed in Ar ambient from 300 to 700 °C for 3 min at each temperature. As the temperature increased, ρc firstly decreased to 4.93 × 10−7 Ω cm2 at 400 °C and then increased. The barrier height was evaluated to be −0.15 eV and − 0.03 eV for as-deposited and 700 °C annealed sample by X-ray photoelectron spectroscopy analysis. Several surface treatments were also carried out to determine their effect on ρc, among which HNO3 vapor treated sample indicates a lower value of 5.32 × 10−6 Ω cm2. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Diamond appears promising for high-power and high-frequency devices, since it has remarkable properties, such as wide band gap (5.47 eV), highest thermal conductivity (22 W/cm K), high breakdown field (N 10 MV/cm), high carrier mobility of electron (4500 cm2 V− 1 s− 1) and hole (3800 cm2 V− 1 s− 1), and high saturation velocity (10 7 cm/s) [1–2]. Recently, applicability of semiconducting diamond films to electronic devices has been extensively studied by fabricating Schottky diodes [3–7] and field-effect transistors (FET) [8–15]. One of the most essential topics to be studied for the diamond applications is the contact properties between the electrode metal layer and the diamond film. Up to now, many refractory metals such as Ti, Mo and Ta, which are capped by Au [12,16,17], are used as contact material for p-type diamond, on account of carbide formation at the interface with the annealing temperature higher than 400 °C. Au [18], a non-carbide-forming metal, was also used as contact metal, whereas the adherence of Au on diamond film is quite poor and could peel off easily even during lift-off process without ultrasonic. However, few investigations on Pd Ohmic contact to hydrogen-terminated single crystal diamond film were reported. ☆ A Publisher’s error resulted in this article appearing in the wrong issue. The article is reprinted here for the continuity of the Special Issue: 9th International Conference on New Diamond and New Carbons (NDNC) 2015. For citation purposes, please use the original publication details; Diamond and Related Materials, Volume 59, pp. 90–94. DOI of original article: http://dx.doi.org/10.1016/j.diamond.2015.09.012. ⁎ Corresponding author. E-mail address: [email protected] (H.X. Wang).
http://dx.doi.org/10.1016/j.diamond.2016.01.019 0925-9635/© 2015 Elsevier B.V. All rights reserved.
In this study, contact properties of Pd on hydrogen terminated diamond surface were investigated. The specific contact resistivity (ρc) was evaluated by Transmission Line Model (TLM). Then, annealing process was performed to examine the thermal stability of the contact. Furthermore, the X-ray photoelectron spectroscopy (XPS) technique was used to identify the contact barrier height at the interface. For comparison, before Pd TLM patterns deposition, the other diamond samples were treated by several kinds of solutions and vapors. 2. Experiment Undoped homoepitaxial diamond films were grown by microwave plasma assisted CVD onto high pressure and high-temperature synthetic Ib diamond (100) substrates with the dimension of 3 × 3 × 0.5 mm3. The total flow rate of the reaction gas was 500 sccm, and the ratio of CH4/H2 was 1%. The process pressure, growth temperature and microwave power were 100 Torr, 900 °C and 1 kW, respectively. After growth, hydrogen plasma was kept for 10 min to form the hydrogen termination, and the samples were cooled down in pure hydrogen ambient. The Hall measurement, Raman spectra, X-Ray Diffraction (XRD) and Atomic Force Microscope (AFM) were carried out to evaluate the epitaxial diamond film. After 3 nm Pd being evaporated on diamond surface, the fabrication process was implemented in two different ways, which is shown in Fig. 1. In one way, 150 nm Pd TLM configuration with an area of 100 × 100 μm2 and space ranging from 5 to 30 μm was firstly formed on diamond film by using the lithography and thermal evaporation method. A negative photoresist (PR) of Az5214 was used, whose prebake and reversal bake temperature were 95 °C and
176
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
Fig. 1. The fabrication process flow of the TLM configuration, a) sample A for annealing process, and b) sample B for surface treatment.
110 °C, respectively. Secondly, the sample was treated by UV/ozone for isolation. Thirdly, the sample was annealed to investigate its thermal stability in Ar (5 N) ambient at several temperatures from 300 to 700 °C during which Ar flow rate was set to 2 L/min and the pressure was kept at atmospheric pressure. Finally, XPS was utilized to determine the barrier height of the Pd contact on hydrogenterminated diamond. This sample was denoted as sample A. In another way, samples were firstly treated by different kinds of solutions and vapors after lithography, including HCl, NH3·H2O and HNO3. Secondly,
150 nm Pd TLM pads were evaporated onto the diamond film. Thirdly, samples were treated by UV/ozone. These samples were named as sample B group. 3. Results and discussion The sample morphologies were investigated by optical microscope (OM) and AFM. The typical images are shown in Fig. 2. Fig. 2(a) shows the full-scale OM image, illustrating a smooth
Fig. 2. a) The optical microscope image of as-deposited single crystal diamond film with the dimension of 3 × 3 mm2, b) AFM image with the area of 5 × 5 μm2, c) the optical microscope image of TLM configuration after annealing at 700 °C in air for 3 min.
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
surface. Fig. 2(b) indicates AFM measurement result, exhibiting a very small surface roughness of 0.318 nm. The mobility, square resistance, carrier density determined by Hall measurement are 92 cm 2 V− 1 s − 1 , 6.5 kΩ, 1.04 × 1013 cm − 2 , respectively. The XRD rocking curve of (400) crystal plane and Raman spectra indicate the good crystalline quality and pure sp 3 structure of the film with relatively low FWHM of 0.01 ° and 2.859 cm− 1, respectively. Fig. 2(c) shows the OM image of sample A after being annealed at 700 °C in Ar atmosphere for 3 min. The edges of all TLM pads are neat, and there is no peeling off phenomenon, indicating good thermal stability. Also, the Pd pads were subjected to ultrasonic during the lift-off process, exhibiting pretty well adherence. In this image, there are some dark features that appeared on the TLM pads, which is ascribed to the scratches induced by the probe. To investigate the thermal stability of the Pd contact on diamond, sample A was annealed by rapid thermal process at several temperatures from 300 to 700 °C in Ar for 3 min. Fig. 3 shows the variation of ρc with the annealing temperature, which is an average value of 6 TLM configurations. As the temperature increases, ρc starts to decrease and reaches the minimum of 4.93 × 10 − 7 Ω cm2 at 400 °C. Compared with ρc of as-deposited sample, the minimum value obtained at 400 °C reduces no more than one order of magnitude, indicating good thermal stability of the Ohmic contact. The initial reduction of ρc may be related with the modification of Pd and diamond interface after annealing. With further increase of the temperature, ρc begins to increase and reaches the maximum at 700 °C with the value of 9.68 × 10− 6 Ω cm 2 , which is larger than that of as-deposited sample. And it's worth noting that the Pd contact on H-terminated diamond can perform excellent Ohmic properties even without annealing. As displayed in Fig. 4(a), the linear fitting curves of the sample annealed at 400 °C and as-deposited sample exhibit a high degree of similarity. The I–V curves in Fig. 4(b) reveal little difference between sample annealed at 700 °C and the as-deposited sample, illustrating no obvious degradation of the Ohmic behavior after such high temperature. Meanwhile, the surface conduction is not reduced, which indicates that the hydrogen termination has been well preserved under such harsh condition. The XPS measurement was used to determine the barrier height of the Pd contact on the diamond film. As shown in Fig. 5, for asdeposited sample, the X-ray beam first focused on the thin Pd to investigate C 1 s and Pd 3d peaks, then X-ray beam was focused on the thick Pd to measure Pd 3d peaks so as to calibrate the C 1s peak of thin Pd. For annealed sample, the X-ray beam first focused on the edge of the Pd pad to investigate C 1s and Pd 3d peaks, then X-ray beam was focused on the Pad to measure Pd 3d peaks so as to calibrate the C 1s peak of Edge. The XPS spectrum of C 1s and Pd 3d are displayed in Fig. 6. Compared with thick Pd and Pad 3d peaks, the 3d peaks of thin Pd and Edge indicate
Fig. 3. The temperature dependence of the specific contact resistance. The annealing temperature is from 300 °C to 700 °C with the atmosphere of Argon for 3 min at each temperature.
177
Fig. 4. a) Linear fit of the same TLM pattern before and after annealing in Ar at 300 °C for 3 min, b) I–V plots of the same two Pd pads before and after annealing in Ar at 700 °C for 3 min.
blue-shift of 0.22 eV and 0.25 eV, respectively. Thus, based on the peak values in Fig. 6(b), the core-level of C 1s is calibrated to be 283.97 − 0.22 = 283.75 eV and 284.13 − 0.25 = 283.88 eV for thin Pd and Edge, respectively. As a result, the barrier height of asdeposited sample is estimated to be −0.15 ± 0.1 eV by using the equation ΦBH = EC-1s-XPS − EC-1s-Rf [19]. The reference binding energy of C 1s (EC-1s-Rf) was determined to be 283.9 ± 0.1 eV with respect to EF [20]. The band diagram of the contact is displayed in Fig. 7. As the barrier height is −0.15 eV, the valance band maximum at the interface bends up. So the holes can easily flow to metal without any barrier, which leads to the excellent Ohmic behavior of Pd contact on the diamond film. Compared with that of as-deposited sample in Fig. 6(a), the Pd 3d peak of annealed sample is scarcely shifted, indicating that there was no carbide formed at the interface of contact. It is one of the reasons that annealing has a little effect on the specific contact resistivity. With the same method mentioned above, the barrier height of 700 °C annealed sample is calculated to be − 0.03 ± 0.1 eV with C 1s peak value of 284.13 eV. The increase of the barrier height gives an explanation why the ρc at 700 °C becomes higher than that of as-deposited. The ρc of sample B group under different surface treatment is revealed in Table 1. Only the treatment of HNO3 vapor can reduce the value of ρc, which could be ascribed to the existence of NO2 in HNO3 atmosphere [21]. However, the other solutions and vapor lead to the degradation of Ohmic properties. One of the possible reason is that
178
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
Fig. 5. The schematic diagram of XPS testing for evaluation of barrier height.
when diamond surface adsorbed HCl and NH3·H2O, the electron is not easily transferred from diamond valance band to adsorbate layer, resulting in low density of 2 dimensional hole gas and lower surface conduction.
4. Conclusion Contact properties of Palladium (Pd) on the surface of hydrogenterminated single crystal diamond have been investigated with several treatment conditions. The ρc of as-deposited sample has been evaluated to be 1.8 × 10−6 Ω cm2. As sample A being annealed from 300 to 700 °C, ρc firstly decreased and then increased. The I–V curves for as-deposited
and annealed sample indicate little difference, illustrating no degradation of Ohmic properties and preserved hydrogen termination. The barrier height of the contact was identified to be − 0.15 ± 0.1 eV and −0.03 ± 0.1 eV for as-deposited and annealed sample by the method of XPS spectra, respectively. And no carbide formation was observed, showing good thermal stability of the contact. Different surface treatments were applied to the sample B and HNO3 vapor was found to enhance the Ohmic properties, the reason may be the existence of NO2 vapor. Prime novelty statement This work further investigates the Pd Ohmic contact properties on intrinsic diamond surface, getting a relatively low specific contact resistance (ρc) of 1.8 × 10−6 Ω cm2. Annealing process with different temperatures indicates the thermal stability of the contact. It's worth noting that the hydrogen termination is preserved after 700 °C annealing. The XPS analysis was carried out to define the barrier heights of as-deposited and annealed samples, explaining the increase of ρc after 700 °C annealing. Also, the surface treatments were applied to verify the effect on ρc.
Fig. 7. The band diagram at the interface of Pd and diamond.
Table 1 The specific Ohmic contact resistance of Pd on single crystal diamond film with different surface treatment. ρc (Ω cm2)
Fig. 6. The XPS spectrums of a) Pd 3d peaks and b) C 1s peaks for as-deposited and annealed samples.
As-deposited HCl NH3·H2O HNO3(vapor) HCl (vapor)
2.14 × 10−5 2.21 × 10−3 2.00 × 10−4 5.35 × 10−6 1.09 × 10−3
W. Wang et al. / Diamond & Related Materials 63 (2016) 175–179
Acknowledgments This paper was supported by the National High-tech R&D Program of China (863 Program) (Grant No.2013AA03A101). References [1] J. Isberg, J. Hammersberg, E. Johansson, T. Wikstrom, D.J. Twitchen, A.J. Whitehead, S.E. Coe, A. Scarsbrook, High carrier mobility in single-crystal plasma-deposited diamond, Science 297 (2002) 1670–1672. [2] L. Reggiani, S. Bosi, C. Canai, F. Nava, Hole-drift velocity in natural diamond, Phys. Rev. B Condens. Matter 23 (1981) 3050–3057. [3] H. Umezawa, S. Shikata, T. Funaki, Diamond Schottky barrier diode for hightemperature, high-power, and fast switching applications, Jpn. J. Appl. Phys. 53 (2014) (05FP06). [4] A. Vescan, I. Daumiller, P. Gluche, W. Ebert, E. Kohn, High temperature, high voltage operation of diamond Schottky diode, Diam. Relat. Mater. 7 (1998) 581–584. [5] D. Takeuchi, S. Yamanaka, H. Okushi, Schottky junction properties of the high conductivity layer of diamond, Diam. Relat. Mater. 11 (2002) 355–358. [6] H. Umezawa, N. Tokuda, M. Ogura, S. Ri, S. Shikata, Characterization of leakage current on diamond Schottky barrier diodes using thermionic-field emission modeling, Diam. Relat. Mater. 15 (2006) 1945–1953. [7] A. Traore, P. Muret, A. Fiori, D. Eon, E. Gheeraert, J. Pernot, Zr/oxidized diamond interface for high power Schottky diodes, Appl. Phys. Lett. 104 (2014) 052105. [8] H. Kawarada, M. Aoki, H. Sasaki, K. Tsugawa, Characterization of hydrogenterminated CVD diamond surfaces and their contact properties, Diam. Relat. Mater. 3 (1994) 961–965. [9] K. Tsugawa, K. Kitatani, H. Noda, A. Hokazono, K. Hirose, M. Tajima, H. Kawarada, High-performance diamond surface-channel field-effect transistors and their operation mechanism, Diam. Relat. Mater. 8 (1999) 927–933. [10] H. Kawarada, High-current metal oxide semiconductor field-effect transistors on H-terminated diamond surfaces and their high-frequency operation, Jpn. J. Appl. Phys. 51 (2012) 090111.
179
[11] A. Daicho, T. Saito, S. Kurihara, A. Hiraiwa, H. Kawarada, High-reliability Passivation of Hydrogen-terminated Diamond Surface by Atomic Layer Deposition of Al2O3, 115 2014, p. 223711. [12] Y. Jingu, K. Hirama, H. Kawarada, Ultrashallow TiC source/drain contacts in diamond MOSFETs formed by hydrogenation-last approach, IEEE Trans. Electron Devices 57 (2010) 966–972. [13] K. Ueda, M. Kasu, Y. Yamauchi, T. Makimoto, M. Schwitters, D.J. Twitchen, G.A. Scarsbrook, S.E. Coe, Diamond FET using high-quality polycrystalline diamond with fT of 45 GHz and fmax of 120 GHz, IEEE Electron Device Lett. 27 (2006) 570–572. [14] K. Hirama, H. Sato, Y. Harada, H. Yamamoto, M. Kasu, Diamond field-effect transistors with 1.3 A/mm drain current density by Al2O3 passivation layer, Jpn. J. Appl. Phys. 51 (2012) 090112. [15] A. Stephen, O. Russell, S. Sharabi, A. Tallaire, D.A.J. Moran, Hydrogen-terminated diamond field-effect transistors with cutoff frequency of 53 GHz, IEEE Electron Device Lett. 33 (2012) 1471–1473. [16] C.M. Zhen, Y.Y. Wang, S.H. He, Q.F. Guo, Z.J. Yan, Y.J. Pu, Ohmic contacts to borondoped diamond, Opt. Mater. 23 (2003) 117–121. [17] J. Nakanishi, A. Otsuki, T. Oku, O. Ishiwata, M. Murakami, Formation of ohmic contacts to P-type diamond using carbide forming metals, J. Appl. Phys. 76 (1994) 2293–2298. [18] C.M. Zhen, X.Q. Wang, X.C. Wu, C.X. Liu, D.L. Hou, Au/p-diamond ohmic contacts deposited by RF sputtering, Appl. Surf. Sci. 255 (2008) 2916–2919. [19] S. Kono, T. Nohara, S. Abe, H. Kodama, K. Suzuki, S. Koizumi, T. Abukawa, A. Sawabe, Electron spectroscopic determination of electronic structures of phosphorus-doped n-type heteroepitaxial diamond (001) surface and junction, Jpn. J. Appl. Phys 51 (2012) 090109. [20] F. Maier, J. Ristein, L. Ley, Electron affinity of plasma-hydrogenated and chemically oxidized diamond (100) surfaces, Phys. Rev. B 64 (2001) 165411. [21] M. Kubovic, M. Kasu, H. Kageshima, Sorption properties of NO2 gas and its strong influence on hole concentration of H-terminated diamond surfaces, Appl. Phys. Lett. 96 (2010) 052101.