Indian Journal of Pure & Applied Physics Vol. 54, June 016, pp. 374-378 Analysis of density and time constant of interface states of MIS device by conductance method A Tataroğlu* & R Ertuğrul Uyar Department of Physics, Faculty of Sciences, azi University, 06500, Ankara, Turkey Received 3 February 016; revised 8 April 016; accepted May 016 The density and time constant of interface states of Au/Si 3 N 4 /n-si (MIS) device have been analyzed by conductance method. The capacitance and conductance measurements of the device have been performed at various frequencies in the range of 1 khz-1 MHz. Experimental results show that p /ω-log(f) plots for each voltage value give a peak because of the presence of interface states. The density (N ss ) and time constant (τ) of interface states have been calculated from ximum value of the peak. The values of N ss and τ range from.49 10 13 ev -1 cm - to 7.57 10 1 ev -1 cm - and from.67 10-5 s to 1.67 10-5 s, respectively. Keywords: MIS device, Conductance method, Interface states, Interface state time constant 1 Introduction The metal-insulator-semiconductor (MIS) and metal-oxide-semiconductor (MOS) devices are capacitor formed from a layer of metal, a layer of insulating or oxide and a layer of semiconductor 1-5. The MIS and MOS devices have an insulating or oxide film thickness which is less than and more than 100 Å, respectively,3. The density of interface states (N ss ) and series resistance (R s ) are significant parameters that cause deviation of the ideal behavior of MOS capacitor 6-9.The interface states are located at or very close to the oxide/semiconductor interface. They cause a bias shift and frequency dispersion of capacitance/conductance-voltage (C/-V) characteristics. Also, the interface states capacitance and resistance are associated with interface states. In order to determine the N ss and R s, there are various methods. Among them, the conductance method is one of the most sensitive methods proposed by Nicollian and oetzberger 10,11. This method is based on measuring the equivalent parallel conductance of MOS device as a function of frequency and bias voltage, and it is used to extract interface states in depletion and cross-section of jority carriers. Interface states having a single trapping time constant (i.e. capture cross-section) will produce a loss peak at a frequency. Moreover, interface traps have different capture cross-section *Corresponding author (E-il: ademta71@gil.com) (loss peak frequency) and surface charge induced potential fluctuations produce a broadening of the loss peak,3,10-1. The purpose of this study is to determine the density and time constant of interface states of Au/Si 3 N 4 /n-si (MIS) device by conductance method. Experimental Details Au/Si 3 N 4 /n-si (MIS) device was fabricated on phosphorus doped (n-type) single crystal Si substrate with a '' diameter, 300 µm thickness, (100) orientation, and 0.5 Ω.cm resistivity. Before the fabrication process, the substrate was chemically cleaned using the conventional method and then chemically etched and finally quenched in deionized water. After cleaning and etching steps, the substrate was mounted on a stainless steel sputtering holder that was heated optically and loaded into a radio frequency gnetron sputtering system. The Si substrate was heated up to 400 C in 10-8 mbar high vacuum and sputter cleaned in pure argon ambient to ensure the removal of any residual organic substance. Then, the silicon nitride (Si 3 N 4 ) film at a constant pressure of 3 10-3 mbar and a constant substrate temperature of 00 C was deposited on the substrate using high-purity (99.999%) silicon nitride target. The ohmic and rectifier contacts were formed using a therl evaporation system. The ohmic back contacts were formed by the deposition of high-pure Au (99.999%) with a thickness of 000 Å at 450 C, under 10-7 mbar vacuum and the sample was annealed
TATAROLU & UYAR : INTERFACE STATES OF MIS DEVICE 375 at 400 C to achieve good ohmic contact behavior. After that, circular dot-shaped rectifier front contacts with mm diameter and 000 Å thickness were formed by the deposition of high-purity Au onto Si 3 N 4 thin film at 50 C. The insulator layer thickness was estited to be about 70 Å. The C/-V measurements were performed by using a HP 419A LF impedance analyzer (5 Hz-13 MHz) in the frequency range of 1 khz-1 MHz at room temperature. 3 Results and Discussion The measured capacitance-voltage (C m -V) and conductance-voltage ( m /ω-v) plots of Au/Si 3 N 4 /n-si (MIS) device as a function of frequency are given in Fig. 1(a-b). As seen in Fig. 1(a-b), the value of both C m and m /ω decrease with increase in the frequency. The frequency dependence of the C and /ω confirms the existence of interface states in the interface of the MIS device. At the lower frequencies, the interface states follow the frequency, while at the higher frequencies they do not follow the frequency of the applied voltage. Therefore, at the sufficiently high frequencies, the contribution of interface states capacitance to the total capacitance can be neglected,3,1-19. Moreover, the C m -V plots give a peak. The peak position shifts towards positive bias voltage with the increasing frequency. This peak can result from R s and the change of N ss. The R s of the MIS and MOS devices can cause a serious error in the extraction of the electrical and dielectric properties. In generally, it can be arisen from ohmic and rectifier contacts, probe wire to the gate, a dirt film or particulate tter between the back contact and the pedestal, the resistance of the quasi-neutral bulk of the semiconductor and an extremely non-uniform doping distribution in the semiconductor. The R s value of the MIS device was calculated using conductance method developed by Nicollian and oetzberger 10,11. According to this method, the measured admittance (Y ) at strong accumulation region of MIS structure using the parallel RC circuit is equivalent to the total circuit admittance as: Fig.1 (a) C m -V and (b) m /ω-v plots of the MIS device as a function of frequency
376 INDIAN J PURE & APPL PHYS, VOL 54, JUNE 016 Y = + iωc (1) where C and are the measured capacitance and conductance in the strong accumulation region, respectively. Series resistance is the real part of the impedance (Z =1/Y ) and is given by the following Eq.: R s = + ( ωc ) () Figure shows the R s -V plots of the MIS device as a function of frequency. As seen in Fig., the calculated R s values decrease with increasing frequency. Such behavior of R s is attributed to the particular distribution of localized interface states. Moreover, the R s -V plots give a peak at about at each frequency. The gnitude of peak increases with decreasing frequency and the peak position shift towards positive bias region (accumulation region) with decreasing frequency due to reordering and restructure of surface states under applied voltage. C m -logf plots of the MIS devices, a function of forward bias voltage, are given in Fig. 3. As seen in Fig. 3, the value of capacitance decreases with the increasing forward bias voltage. After a certain value of the frequency, the measured capacitance decreases with increase in the frequency. m /ω-logf plots of the MIS device are given in Fig. 4. As seen in Fig. 4, the value of conductance increases with the increasing forward bias voltage, while it decreases with the increasing frequency. The frequency dependence of the capacitance and conductance is attributed to the presence of a continuous distribution of interface states 0-5. The conductance method developed by Nicollian and oetzberger is the most sensitive method for determining the density and time constant of interface states,10,11. The conductance method is based on measuring the equivalent parallel conductance as a function of bias voltage and frequency. According to this method, the equivalent parallel conductance ( p ) can be expressed as 11,6-31 : ] p qnss = ln(1 + ω ω ωτ τ ) (3) where ω is the angular frequency and τ is the time constant of the interface states (i.e. capture crosssection). In practice, different interface traps have different capture cross-section (loss peak frequency) and surface charge induced potential fluctuations produce a broadening of the loss peak. The equivalent Fig. 3 C m -logf plots of the MIS device as a function of forward bias voltage Fig. Series resistance-voltage (R s -V) plots of the MIS device as a function of frequency Fig. 4 m /ω-logf plots of the MIS device as a function of forward bias voltage
TATAROLU & UYAR : INTERFACE STATES OF MIS DEVICE 377 parallel conductance related to the measured impedance by,11, Fig. 5 p /ω-logf plots as a function of forward bias voltage Table 1 Calculated values of the density and time constant of interface states. V (V) N ss (ev -1.cm - ) τ (s) 1.0.49 10 13.67 10-5 1.5 1.89 10 13.7 10-5.0 1.44 10 13.06 10-5.5 1.18 10 13 1.93 10-5 3.0 1.04 10 13 1.84 10-5 3.5 9.57 10 1 1.79 10-5 4.0 9.06 10 1 1.75 10-5 4.5 8.70 10 1 1.7 10-5 5.0 8.47 10 1 1.71 10-5 5.5 7.64 10 1 1.68 10-5 6.0 7.57 10 1 1.67 10-5 ω C p m ox = ω m + ω ( Cox Cm) (4) The values of the parallel conductance were obtained by using Eq. (4). The p /ω plots as a function of forward bias voltage are given in Fig. 5. As seen in Fig. 5, the p /ω curves for each forward bias voltage show a peak. It is clear that the peak position shifts towards higher frequencies with increase in the forward bias voltage. The gnitude of peak depends on the capture rate, that is, the interface state level occupancy that is determined by the applied bias voltage. When the ac signal corresponds to this time constant, the loss peak associated to the interface trap levels will occur 1,3-37. Maximum loss peak occurs when interface traps are in resonance with the applied ac signal. At this peak [ ( p /ω)]/ (ωτ)=0 and this ximum condition gives ωτ=1.98. The interface state time constant is calculated from τ=1.98/ω, where ω is the frequency at which the p /ω peak occurs. The N ss can be obtained from the parallel conductance peak using the following equation: N ss ( p / ω) x = (5) 0.40qA The calculated values of the N ss and τ of the MIS device are given in Table 1. As seen in Table 1, the values of both N ss and τ decrease with increase in the forward bias voltage and are changed from.49 10 13 ev -1 cm - to 7.57 10 1 ev -1 cm - and from.67 10-5 s to 1.67 10-5 s, respectively. 4 Conclusions In this study, the density and time constant of interface states of the fabricated MIS device were analyzed by capacitance-conductance-voltage measurements. The experimental results indicate that both the measured capacitance and conductance vary with applied bias voltage and frequency. The parallel conductance curves for each forward bias voltage value give a peak. The values of the N ss and τ calculated from the ximum value of the parallel conductance peak decrease with the increasing forward bias voltage. The obtained results indicate that the interface states affect strongly the electrical characteristics of the MIS device. References 1 Bentarzi H, Transport in Metal-Oxide-Semiconductor Structures (Springer, New York), 011. Nicollian E H & Brews J R, MOS Physics and Technology (Wiley, New York), 198. 3 Sze S M, Physics of Semiconductor Devices (Wiley, New York), 1981. 4 Chu J & Sher A, Device Physics of Narrow ap Semiconductors (Springer, New York), 010. 5 Moslehi M, IEEE TransElectron Device, ED-3 (1985) 106. 6 Ertuğrul R & Tataroğlu A, Rad Effects Defects Solids, 169 (014) 791. 7 Tataroglu A & Altindal S, Microelectron Eng, 85 (008) 56. 8 Tataroglu A, Altındal S & Bulbul M M, Microelectron Eng, 81 (005) 140. 9 Kar S & Dahlke W E, Solid State Electron, 15 (197) 1. 10 Nicollian E H & oetzberger A, Appl Phys Lett, 7 (1965) 16. 11 Nicollian E H & oetzberger A, Bell Syst Tech J, 46 (1967) 1055. 1 Singh J, Semiconductor Devices, Basic Principles (Wiley, New York), 001.
378 INDIAN J PURE & APPL PHYS, VOL 54, JUNE 016 13 Baran H M & Tataroglu A, Chin Phys B, (013) 047303. 14 Kim H, Kim H & Kim D W, Vacuum, 101 (014) 9. 15 Pratap Reddy M S, Lee J H & Jang J S, Synth Met, 185-186 (013) 167. 16 Altındal S, Yucedag I & Tataroglu A, Vacuum, 84 (009) 363. 17 Fouad S S, Sakr B, Yahia I S, Abdel-Basset D M & Yakuphanoğlu F, Mater Res Bull, 49 (014) 369. 18 Hussain I, Soomro M Y, Bano N, Nur O & Willander M, J Appl Phys, 11 (01) 064506. 19 KayaS, Lok R, Aktag A, Seidel J & Yılz E, J AlloysCompd, 583 (014) 476. 0 Dokme I & Altindal S, Physica B, 391 (007) 59. 1 Pad R, Lakshmi B P & Reddy V R, Superlattices Microstruc, 60 (013) 358. Baris B, Physica B, 46 (013)13. 3 Reddy V R, Thin Solid Films, 556 (014) 300. 4 Tataroglu A & Altındal S, Vacuum, 8 (008) 103. 5 Aydın H, Tataroğlu A, Al-hamdi A A, Yakuphanoglu F, El-Tantawy F & Farooq W A, J Alloys Compd, 65 (015) 18. 6 Yang H, Son Y, Choi S & Hwang H, Jpn J ApplPhys, 44 (005) 1460. 7 Engel-Herbert R, Hwang Y & Stemmer S, J Appl Phys, 108 (010) 14101. 8 Aydın M E, Yakuphanoğlu F & Öztürk, Synth Metals, 160 (010) 186. 9 Chattopadhyay P & Daw A N, Solid-State Electron, 9 (1986) 555. 30 Tascıoglu I, Soylu M, Altındal S, Al-hamdi A A & Yakuphanoglu F, J Alloys Compd, 541 (01) 46. 31 Miller E J, Dang X Z, Wieder H H, Asbeck P M,Yu E T, Sullivan J & Redwing J M, J Appl Phys, 87 (000) 8070. 3 Yun M, angopadhyay S, Bai M, Taub H, Arif M & uha S, Organic Electron, 8 (007) 591. 33 Wong H & ritsenkov A, Microelectron Reliab, 4 (00) 597. 34 Bülbül M M, Altındal S, Parlaktürk F & Tataroğlu A, Surf Interface Anal, 43 (011) 1561. 35 DinaraS M, hosh S, Halder N N, Bag A, Bhattacharya S & Biswas D, Microelectron Reliab, 55 (015) 789. 36 Werner J, Ploog K & Queisser H J, Phys Rev Lett, 57 (1986) 1080. 37 Chakraborty S, Bera M K, Bose P K & Maitii C K, Semicond Sci Technol, 1 (006) 335.