Capacitance-voltage characteristics

Figure 4.2 shows the multi-frequency C-V responses and quasi-static C-V (QSCV) curves of 18 nm Al₂O₃/n-InAs MOSCAP samples. The multi-frequency C-V measurement was made using an HP4284A LCR meter, and QSCV curves were acquired using an Agilent precision 4156C analyzer. Measured leakage currents were smaller than 10^-9 A/cm² in the range ± 3.5 V gate bias for all samples, ensuring they did not influence the QSCV measurements (see Fig. 4.3). In the accumulation regime, the multi-frequency responses do not show the obvious difference in frequency dispersion between samples. As shown in Figs. 4.2a - 2c, the values of frequency dispersion of samples are small, in the range of 0.65-0.75% per decade. These low frequency dispersions including the control sample indicate that surface treatments do not seem to affect significantly the C-V responses in the accumulation regime.

Figure 4.2. Multi-frequency C-V responses (solid lines) and QSCV curves (dashed lines) in a- control sample, b- HCl plus TMA sample, and c- sulfide plus TMA treated sample of 18 nm ALD Al2O3/InAs MOSCAPs; d- QSCV curves of all three samples, for comparison

68

Figure 4.3. Leakage currents versus gate voltage of samples

Low frequency-like C-V behavior is observed for all samples in the whole range of measured frequencies. This behavior originates from the short minority carrier response time (_R) in very low band gap, high intrinsic density material as InAs. As shown in Figs.

4.2a and 2d, the control sample exhibits high value of depletion capacitance (Cdep) in depletion regime which reveals large value of interface trap capacitance (C_it). Large frequency dispersion in inversion regime in this sample as shown in Fig. 4.2a also implies high contribution of interface traps. In chemicals plus TMA treated samples, nice C-V curves with small frequency dispersion in inversion regime are observed (Figs. 4.2b - 2c).

In Fig. 4.2d, smaller C_dep of these two samples compared to the control sample indicates that the contribution of C_it is reduced.

Out of the two chemical plus TMA treated samples, the HCl plus TMA sample exhibits better electrical characteristics as compared to the sulfide plus TMA treated sample, with smaller frequency dispersion in inversion regime and smaller stretch out (Figs. 4.2b-2d). This result seems contradictory to most of reports on high k/ GaAs (InGaAs) but it is consistent with the report on HfO2/InAs structure [7]. Moreover, previous report had shown that during thermal process sulfur atoms could diffuse into InAs [14], resulting in the change of doping concentration at ultrathin InAs layer near high k/InAs interface. This change may also influence the electrical properties of the sulfide plus TMA treated sample.

69 4.3.3. Simulation and D_it profiles extraction

Low frequency CV-simulations were performed by full numerical solution of the Poisson equation: used is one-dimensional using simple finite difference discretization and Cauchy boundary conditions at the semiconductor-oxide interface. The full heterostructure of the device has been taken into account. At the InAs top surface, the charge due to interface states is taken into account as well, similar to the approach in [13]. The interface state density (Dit) at the InAs/high-k interface was varied, until a good fit to the experimental data was obtained. For the electron density approximation, a model using a correction for the non-parabolicity of the conduction band was used [15]:

√ ∫

(4.2)

where,  = (E-EC)/kT is the normalized electron kinetic energy, = (EF-E_C)/kT is the reduced Fermi energy, N_C is the effective density of states in the conduction band, and α is the nonparabolicity factor:

( ) (4.2)

here, me is the electron effective mass, m0 is the free electron mass and g = (EC-EV)/kT is the normalized bandgap.

All experimental QSCV curves (symbols) were well fitted by the simulations (solid lines). Dit profiles of samples extracted from simulation are shown in Figs. 4.4b-4d, where the estimated error bars of the extracted Dit are shown as well. Errors were estimated and taken into account due to the following reasons: (i) error on metal work function, (ii) charge quantization effects which were not included in the simulation and (iii) uncertainty

70

on absolute value of oxide capacitance, C_ox. The derived D_it profiles present a U-shape with minimum in Dit profile located around the conduction band minimum (EC) for all samples. The interface state density shows strong similarities with the In_0.53Ga_0.47As/Al₂O₃ D_it profile [13]. It can be clearly seen that the two different surface treatments significantly reduce the donor-like traps over the full energy region as compared to the control sample (Figs. 4.4b-4d). The D_it of the sulfide treated plus TMA sample shows slightly higher values of donor-like traps as compared to the HCl plus TMA treated sample, as expected from the general C-V characteristics of these two samples with deeper depletion of the HCl treated sample as compared to the (NH4)2S treated sample.

Figure 4.4. a- Experimental data (symbols), simulated C-V curves (solid lines) of ALD 18 nm Al2O3/n-InAs MOSCAP samples with various surface treatments. Interface state density profiles of all three samples, extracted from simulation, are shown as well: b- control sample, c- HCl plus TMA treated sample, d- sulfide plus TMA treated sample.

4.4. Conclusions

We have examined the effect of surface treatments on the physical and electrical properties of the Al2O3/n-InAs structures. The effect of interface states on accumulation

71

capacitance behavior is small and does not depend on the surface treatments. In contrast, the C-V characteristics of Al2O3/n-InAs in depletion and inversion region were significantly improved by surface treatments. The D_it profiles extracted from simulation shows a significant reduction of donor-like traps after surface treatments in complete InAs bandgap, as well as in the lower part of conduction band. Results also revealed that HCl plus TMA treatment has stronger effect on the reduction of donor-like traps than sulfide plus TMA treatment.

72 References

[1] A. P. Kirk, M. Milojevic, J. Kim, and R. M. Wallacea, "An in situ examination of atomic layer deposited alumina/InAs(100) interfaces," Appl. Phys. Lett. 96, 202905 (2010).

[2] R. Timm, A. Fian, M. Hjort, C. Thelander, E. Lind, J. N. Andersen, L.-E. Wernersson, and A. Mikkelsen, "Reduction of native oxides on InAs by atomic layer deposited Al2O3 and HfO2," Appl. Phys. Lett. 97, 132904 (2010).

[3] Abigail Lubow, Sohrab Ismail-Beigi, and T. P. Ma, "Comparison of drive currents in metal-oxide-semiconductor field-effect transistors made of Si, Ge, GaAs, InGaAs, and InAs channels," Appl. Phys. Lett. 96, 122105 (2010).

[4] Erik Lind, Yann-Michel Niquet, Hector Mera, and Lars-Erik Wernersson, "Accumulation capacitance of narrow band gap metal-oxide-semiconductor capacitors," Appl. Phys. Lett.

96, 233507 (2010).

[5] Hai-Dang Trinh, Edward Yi Chang, Yuen-Yee Wong, Chih-Chieh Yu, Chia-Yuan Chang, Yueh-Chin Lin, Hong-Quan Nguyen, and Binh-Tinh Tran "Effects of Wet Chemical and Trimethyl Aluminum Treatments on the Interface Properties in Atomic Layer Deposition of Al2O3 on InAs," Jpn. J. Appl. Phys. 49, 111201 (2010).

[6] Hyoung-Sub Kim, I. Ok, M. Zhang, F. Zhu, S. Park, J. Yum, H. Zhao, Jack C. Lee, Prashant Majhi, N. Goel, W. Tsai, C. K. Gaspe, and M. B. Santos, "A study of metal-oxide-semiconductor capacitors on GaAs, In0.53Ga0.47As, InAs, and InSb substrates using a germanium interfacial passivation layer," Appl. Phys. Lett. 93, 062111 (2008).

[7] D. Wheeler, L.-E. Wernersson, L. Fröberg, C. Thelander, A. Mikkelsen, K.-J. Weststrate, A.

Sonnet, E.M. Vogel, and A. Seabaugh, "Deposition of HfO2 on InAs by atomic-layer deposition," Microelectron. Eng. 86, 1561-1563 (2009).

[8] Yun-Chi Wua, Edward Yi Chang, Yueh-Chin Lin, Chi-Chung Kei, Mantu K. Hudait, Marko Radosavljevic, Yuen-Yee Wong, Chia-Ta Chang, Jui-Chien Huang, and Shih-Hsuan Tang, "Study of the inversion behaviors of Al2O3/InxGa1-xAs metal-oxide-semiconductor capacitors with different In contents," Solid-State Electron. 54, 37-41 (2010).

[9] Ning Li, Eric S. Harmon, James Hyland, David B. Salzman, T. P. Ma, Yi Xuan, and P. D.

Ye, "Properties of InAs metal-oxide-semiconductor structures with atomic-layer-deposited Al2O3 Dielectric," Appl. Phys. Lett. 92, 143507 (2008).

[10] Hyunhyub Ko, Kuniharu Takei, Rehan Kapadia, Steven Chuang, Hui Fang, Paul W. Leu, Kartik Ganapathi, Elena Plis, Ha Sul Kim, Szu-Ying Chen, Morten Madsen, Alexandra C.

Ford, Yu-Lun Chueh, Sanjay Krishna, Sayeef Salahuddin, and Ali Javey, "Ultrathin compound semiconductor on insulator layers for high-performance nanoscale transistors,"

Nature 468, 286-289 (2010).

73

[11] M. Milojevic, C. L. Hinkle, F. S. Aguirre-Tostado, H. C. Kim, E. M. Vogel, J. Kim,nand R.

M. Wallace, "Half-cycle atomic layer deposition reaction studies of Al2O3 on (NH4)2S passivated GaAs(100) surfaces," Appl. Phys. Lett. 93, 252905 (2008).

[12] H. D. Trinh, E. Y. Chang, P. W. Wu, Y. Y. Wong, C. T. Chang, Y. F. Hsieh, C. C. Yu, H. Q.

Nguyen, Y. C. Lin, K. L. Lin, and M. K. Hudait, "The influences of surface treatment and gas annealing conditions on the inversion behaviors of the atomic-layer-deposition Al2O3 /n-In0.53Ga0.47As metal-oxide-semiconductor capacitor," Appl. Phys. Lett. 97, 042903 (2010).

[13] G. Brammertz, H.-C. Lin, M. Caymax, M. Meuris, M. Heyns, and M. Passlack, "On the interface state density at In0.53Ga0.47As/oxide interfaces," Appl. Phys. Lett. 95, 202109 (2009).

[14] Johnny C. Ho, Alexandra C. Ford, Yu-Lun Chueh, Paul W. Leu, Onur Ergen, Kuniharu Takei, Gregory Smith, Prashant Majhi, Joseph Bennett, and Ali Javey, "Nanoscale doping of InAs via sulfur monolayers," Appl. Phys. Lett. 95, 072108 (2009).

[15] Viktor Ariel-Altschul, Eliezer Finkman, and Gad Bahir, "Approximations for carrier density in nonparabolic semiconductors," IEEE Trans. Electron Devices 39, 1312-1316 (1992).

74

Chapter 5

T