The improvement to the reliability by different concentration of ICP N 2

plasma nitridation

Fig. 4-17 shows the results from a constant-current stress experiment of oxynitride films that nitrided by different ICP plasma of 200 W at 5 min. The stress current was -100 mA/cm². The improvement to constant-current stress characteristic by ICP N₂ mixed inert gas plasma nitridation was better than pure N₂ plasma nitridation. And the improvement of N2 mixed He plasma nitridation was the best.

Fig. 4-18 shows the Stress-induced-leakage-current (SILC) of oxynitride films that nitrided by different ICP plasma of 200 W at 5 min. The condition of stress was on -1 V at 10 min. After ICP N₂ mixed inert gas plasma nitridation, the SILC of oxynitride may be negligible.

Charge to breakdown of oxynitride films that nitrided by different ICP plasma of 200 W at 5 min shows in fig. 4-19. The characteristic of charge to breakdown was

measured at constant-current of -1 A/cm². The oxynitride nitrided by using ICP N₂ mixed He plasma of 200 W also had the best performance on charge to breakdown.

From above results, we may conclude the best method to improve the reliabilities of oxynitride was by ICP N2 mixed He plasma of 200 W at 5 min.

4.3 Summary

After developing an acceptable method to grow ultrathin oxynitride, we tried to use HD-ICP plasma nitridation to improve oxynitride to get better performance on electrical properties and reliabilities. First, we tried to find the most suitable process time of ICP N₂ plasma nitridation to get best electrical properties. Then we used the method of ICP N2 plasma nitridation to oxynitride or silicon dioxide grown by different ICP plasma for verifing this nitridation technology. Furthermore, we changed the pure N2 plasma to N2 mixed inert gas plasma for enhance nitridation effect. And we discussed the performance of ICP N₂ plasma nitridation to reliability. We had succeeded to find out an applicable way to enhance original oxynitride grown by ICP plasma. Both the electrical properties and reliability could get improvement by ICP N₂ mixed He plasma nitridation.

4.4 References

[1]S. V. Hattangady, R. Kraft, D. T. Grider, M. A. Douglas, G. A. Brown, P. A. Tiner, J. W.

Kuehne, P. E. Nicollian, and M. F. Pas, “Ultrathin nitrogen-profile engineered gate dielectric films,” in IEDM Tech. Dig., Dec. 1996, pp. 495-498.

[2]S. F. Ting, Y. K. Fang, C. H. Chen, C. W. Yang, W. T. Hsieh, J. J. Ho, M. C. Yu, S. M.

Jang, C. H. Yu, M. S. Liang, S. Chen, and R. Shih, “The effect of remote plasma nitridation on the integrity of the ultrathin gate dielectric films in 0.13 µm CMOS technology and beyond,” IEEE Electron Device Lett., 2001, 22, pp. 327-329.

[3]S. F. Ting, Y. K. Fang, C. H. Chen, C. W. Yang, W. T. Hsieh, J. J. Ho, M. C. Yu, S. M.

Jang, C. H. Yu, M. S. Liang, S. Chen, and R. Shih, “The effect of remote plasma nitridation on the integrity of the ultrathin gate dielectric films in 0.13 μm CMOS technology and beyond,” IEEE Electron Device Lett., vol. 22, pp. 327–329,July 2001.

[4]D. Ishikawa, S. Sakai, K. Katsuyama, and A. Hiraiwa, “Nitride-sandwiched-oxide gate insulator for low power CMOS,” in IEDM Tech. Dig., Dec., 2002, pp. 869-872.

[5]Chien-Hao Chen; Yean-Kuen Fang; Shyh-Fann Ting; Wen-Tse Hsieh; Chih-Wei Yang;

Tzu-Hsuan Hsu; Mo-Chiun Yu; Tze-Liang Lee; Shih-Chang Chen; Chen-Hua Yu; and Mong-Song Liang; “Downscaling limit of equivalent oxide thickness in formation of ultrathin gate dielectric by thermal-enhanced remote plasma nitridation,” IEEE Trans.

Electron Devices, vol. 49, pp. 840-846, May, 2002.

[6]Y. Saito, K. Sekine, M. Hirayama, and T. Ohmi, “Low-temperature formation of silicon nitride film by direct nitridation employing high-density and low-energy ion bombardment,” Jpn. J. Appl. Phys., pt. 1, vol. 38, no. 4, pp. 418-422, 1999.

[7]A. A. Bright, J. Batey, and E. Tierny, “Low-rate plasma oxidation of Si in a dulite oxgen/helium plasma for low-temperature gate quality Si/SiO₂ interfaces,” Appl. Phys.

Lett., vol. 11, pp. 619-621, 1991.

Chapter 5 Conclusions

In this thesis, we have successfully demonstrated the growth of ultrathin oxynitride films at a low process temperature of 300 ℃ by using an inductively coupled high-density plasma source. The RF power and gas source of plasma were optimized to have enough nitrogen both for suppressing the growth rate and forming strong silicon-nitrogen bonds. The uniformity and the smoothness at the oxynitride/silicon interface of the ICP plasma ultrathin oxynitride film are acceptable.

The experimental results showed that 1.0nm oxynitride grown by ICP N2O plasma of 200 W at 300℃ for 1 min has the highest capacitance and the lowest leakage current density of -0.1 A/cm² at -1 V. In the direct tunneling regime, a higher nitrogen concentration leads to less leaky film. Nitridation decreases the effective dielectric thickness by increase the dielectric constant of the film. In our study, we have developed high-quality silicon oxynitride gate dielectric with physical thickness of 1.0 nm. The boron penetration was suppressed by increasing nitrogen concentration more than 1 x 10²¹ cm^-3 even with ultrathin base oxide. In our experiments, the nitrogen concentration of the oxynitride surface was about 8 x 10²¹ cm^-3. The boron penetration would be suppressed because of higher nitrogen concentration incorporated in the oxynitride surface.

We have also examined the effect of ICP N₂ plasma nitridation on the characteristics of oxynitrides grown by ICP plasma. Proper nitridation to the oxynitride could reduce the fixed charge density and suppress electron trapping. The experimental results showed that leakage current density of 1.0 nm oxynitride grown

by ICP N₂O plasma of 200 W at 300 ℃ for 1 min following by ICP N₂ mixed He plasma of 200 W at 300 ℃ for 5 min nitridation is lower than one without plasma nitridation. And the capacitance density and reliability of ultrathin oxynitride are also improved by plasma nitridation. We also have examined that more nitrogen could be incorporated into the base oxynitride effectively without increasing the final physical oxide thickness obviously by ICP N2 plasma nitridation technology.

Finally, we have successfully achieved a low leakage and high quality 1.0 nm oxynitride film at a low process temperature as 300 ℃. This 1.0 nm thick oxynitride film is suitable to be applied as gate dielectric of next-generation high-performance 65 nm MOSFET devices.

Pure N₂O ambient, gas flow = 100 sccm

Plasma power = 200 W, Oxidation time = 1 min, Temperature = 300 ℃

Pressure(mTorr) 10 30 50 100 150

Thickness(Å) 8 10.2 10 10 10.4

N₂+O₂ ambient, N₂ gas flow = 100 sccm O₂ gas flow = 20 sccm

Plasma power = 200 W, Oxidation time = 1 min, Temperature = 300 ℃

Pressure(mTorr) 10 30 50 100 150

Thickness(Å) 7.5 8.3 8.5 8.5 8.5

Pure O2 ambient, gas flow = 100 sccm

Plasma power = 200 W, Oxidation time = 1 min, Temperature = 300 ℃

Pressure(mTorr) 10 30 50 100 150

Thickness(Å) 10 11.3 11.5 12 13

Table3-1

The thickness of oxynitride films versus working pressure.

Pure N₂O ambient, gas flow = 100 sccm

Oxidation time = 1 min, Temperature = 300 ℃, Working pressure = 100 mTorr

Power 30 50 100 200 300 400 600

Thickness(Å) 8.5 9.2 9.5 10 10.4 12 15

N₂+O₂ ambient, N₂ gas flow = 100 sccm O₂ gas flow = 20 sccm

Oxidation time = 1 min, Temperature = 300 ℃, Working pressure = 100 mTorr

Power 30 50 100 200 300 400 600

Thickness(Å) 7.8 8 8.3 8.5 9.8 11.5 13

Pure O2 ambient, gas flow = 100 sccm

Oxidation time = 1 min, Temperature = 300 ℃, Working pressure = 100 mTorr

Power 30 50 100 200 300 400 600

Thickness(Å) 10.5 10.5 11.2 12 14 15 18

Table3-2

The thickness of oxynitride versus plasma power

Pure N₂O ambient, gas flow = 100 sccm

Plasma power = 200 W, Temperature = 300 ℃, Working pressure = 100 mTorr

Time(min) 0.5 1 2 3 5 10

Thickness(Å) 8 10 11 12.4 14.8 18.4

N₂+O₂ ambient, N₂ gas flow = 100 sccm O₂ gas flow = 20 sccm

Plasma power = 200 W, Temperature = 300 ℃, Working pressure = 100 mTorr

Time(min) 0.5 1 2 3 5 10

Thickness(Å) 8 8.5 9 10.8 13.4 15

Pure O₂ ambient, gas flow = 100 sccm

Plasma power = 200 W, Temperature = 300 ℃, Working pressure = 100 mTorr

Time(min) 0.5 1 2 3 5 10

Thickness(Å) 10 12 13 15 17 22

Table 3-3

The thickness of oxynitride versus processing time

Fig. 2-1 The ICP system that was used in this experiment.

Fig. 2-2 The Experimental Procedure of ICP Plasma Oxynitridation Process.

(a)RCA clean

(b)Plasma oxynitridation

(c)Deposit polysilicon

(d)Ion implantation

(e)Ni silicide

(f)Deposit Al

(g)Pattern upper electrode

(h)Back electrode

Fig.2-3 The detail fabrication process flow of MOS Capacitor with 1.0 nm oxynitride

Fig 2-4 Principles of an ellipsometric measurement

Fig 2-5 Spectroscopic ellipsometry measured SiO2 film thickness vs. thickness measured by, XPS, C–V analysis, and TEM. Reference: “SiO2 film thickness metrology by x-ray photoelectron spectroscopy,” Appl. Phys. Lett., vol. 71, pp.

(a)

(b)

Fig 2-6 Schematic illustration of (a) FN-tunneling and (b) direct tunneling mechanisms of electron flow through an oxide potential barrier of height ΦB.

Fig. 3-1 The dependence of oxide thickness on chamber pressure.

Fig. 3-2 The dependence of oxide thickness on RF power.

Fig. 3-3 The dependence of oxide thickness on oxidation time.

Fig. 3-4 High-resolution cross-sectional TEM photograph of MOS capacitor with 1.0 nm thick oxynitride gate dielectric. The capacitor was formed by depositing 100 nm ploy-Si/1 nm oxsynitride on Si substrate.

Fig. 3-5 Enlarged high-resolution TEM micrographs of 1 nm oxynitride sample for the polysilicon-oxynitride-silicon structure.

Fig. 3-6 The high-frequency (100 kHz) C-V curves of MOS capacitors with 1.0 nm gate dielectrics grown by HD-ICP N₂O-plasma at various RF powers.

Fig. 3-7 The current density versus gate voltage (J_g-V_g) characteristics of the 1.0 nm gate dielectrics grown by HD-ICP N2O-plasma at various RF powers.

Fig. 3-8 The high-frequency (100 kHz) C-V curves of MOS capacitors with 1.0 nm gate dielectrics grown by HD-ICP N₂O, O₂, and N₂/O₂-mixture-plasma at RF power of 200 W, respectively.

Fig. 3-9 The current density versus gate voltage (J_g-V_g) characteristics of the 1.0 nm gate dielectrics grown by HD-ICP N2O, O2, and N2/O2-mixture-plasma at RF power of 200 W, respectively.

Fig. 3-10 The SIMS nitrogen profiles of the 1.0 nm gate dielectric films of the 1.0 nm gate dielectrics grown by HD-ICP N₂O and N₂/O₂-mixture-plasma at RF power of 200 W, respectively.

Fig. 3-11 The charge trapping characteristics by monitoring the change in gate voltage (ΔVg) as a function of stress time for 1.0 nm oxynitride films grown by HD-ICP N₂O-plasma at various RF powers.

Fig. 3-12 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films grown by HD-ICP N2O-plasma at various RF powers.

Fig. 3-13 Charge-to-breakdown characteristics (Q_bd) of the oxynitride films grown by HD-ICP N2O-plasma at various RF powers under constant current stress (J = -1 A/cm²).

Fig. 3-14 The charge trapping characteristics by monitoring the change in gate voltage (ΔVg) as a function of stress time for 1.0 nm oxynitride films grown by HD-ICP N2O, O2, and N2/O2-mixture-plasma at RF power of 200 W, respectively.

Fig. 3-15 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films grown by HD-ICP N2O, O2, and N2/O2-mixture-plasma at RF power of 200 W, respectively.

Fig. 3-16 Charge-to-breakdown characteristics (Q_bd) of the oxynitride films grown by HD-ICP N₂O, O₂, and N₂/O₂-mixture-plasma at RF power of 200 W, respectively under constant current stress (J = -1 A/cm²).

Fig. 4-1 The experimental procedure of HD-ICP nitridation-plasma process

Fig. 4-2 The high-frequency (100 kHz) C-V curves of MOS capacitors with 1.0 nm gate dielectrics nitrided by HD-ICP N2-plasma with different nitridation-time.

Fig. 4-3 The current density versus gate voltage (Jg-Vg) characteristics of the 1.0 nm gate dielectrics nitrided by HD-ICP N2-plasma with different nitridation-time.

Fig. 4-4 The SIMS nitrogen profiles of the 1.0 nm gate dielectric films of the 1.0 nm gate dielectrics nitrided by HD-ICP N₂-plasma with different nitridation-time.

Fig. 4-5 The high-frequency (100 kHz) C-V curves of MOS capacitors with the 1.0 nm gate dielectrics grown by HD-ICP N2O, O2, and N2/O2-mixture-plasma with and without nitridation, respectively.

Fig. 4-6 The current density versus gate voltage (J_g-V_g) characteristics of 1.0 nm gate dielectrics grown by HD-ICP N₂O, O₂, and N₂/O₂-mixture-plasma with and without nitridation, respectively.

Fig. 4-7 The high-frequency (100 kHz) C-V curves of MOS capacitors with the 1.0 nm gate dielectrics nitrided by HD-ICP N2O, N2/Ar-mixture, and N2 /He-mixture-plasma, respectively.

Fig. 4-8 The current density versus gate voltage (J_g-V_g) characteristics of the 1.0 nm gate dielectrics nitrided by HD-ICP N2O, N2/Ar-mixture, and N2/He-mixture-plasma, respectively.

Fig. 4-9 The charge trapping characteristics by monitoring the change in gate voltage (ΔVg) as a function of stress time for 1.0 nm oxynitride films nitrided by HD-ICP N2 -plasma with different nitridation-time.

Fig. 4-10 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films nitrided by HD-ICP N2-plasma with different nitridation-time.

Fig. 4-11 Charge-to-breakdown characteristics (Q_bd) of the oxynitride films nitrided by HD-ICP N2-plasma with different nitridation-time.

Fig. 4-12 The charge trapping characteristics by monitoring the change in gate voltage (ΔVg) as a function of stress time for 1.0 nm oxynitride films grown by HD-ICP N2O, O2, and N2/O2-mixture-plasma with and without nitridation, respectively.

Fig. 4-13 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films grown by HD-ICP N2O-plasma with and without nitridation.

Fig. 4-14 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films grown by HD-ICP O₂-plasma with and without nitridation.

Fig. 4-15 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films grown by HD-ICP N2/O2-mixture-plasma with and without nitridation.

Fig. 4-16 Charge-to-breakdown characteristics (Qbd) of the oxynitride films grown by HD-ICP N₂O, O₂, and N₂/O₂-mixture-plasma with and without nitridation, respectively.

Fig. 4-17 The charge trapping characteristics by monitoring the change in gate voltage (ΔVg) as a function of stress time for 1.0 nm oxynitride films nitrided by HD-ICP N₂O, N₂/Ar-mixture, and N₂/He-mixture-plasma, respectively.

Fig. 4-18 Stress-induced leakage current (SILC) of capacitors with 1.0 nm oxynitride films nitrided by HD-ICP N2O, N2/Ar-mixture, and N2/He-mixture-plasma, respectively.

Fig. 4-19 Charge-to-breakdown characteristics (Q_bd) of the oxynitride films nitrided by HD-ICP N2O, N2/Ar-mixture, and N2/He-mixture-plasma, respectively.