Academia.eduAcademia.edu

Figure 9 – uploaded by GEORGE PAPAIOANNOU

See full PDFdownloadDownload figure
Fig. 8. The stored charge, oo calculated from TSDC measurements versus temper- ature for HF SiN samples.

Figure 8 The stored charge, oo calculated from TSDC measurements versus temper- ature for HF SiN samples.

Related Figures (10)

Investigated test structures. Table 1
Investigated test structures. Table 1
Fig. 1. Time dependence of the transient current for HF SiN with 400 nm thickness under positive bias: (a) charging current and (b) discharging current.
Fig. 1. Time dependence of the transient current for HF SiN with 400 nm thickness under positive bias: (a) charging current and (b) discharging current.
Fig. 2. The absolute value of charging and discharging current transients at E=140V/um for HF SiN samples with 400 nm thickness.
Fig. 2. The absolute value of charging and discharging current transients at E=140V/um for HF SiN samples with 400 nm thickness.
Fig. 3. The decay time constant, t versus electric field intensity in HF SiN wit different thicknesses. Another contribution to the increase of the decay time constant with the electric filed intensity arises from the tunneled electrons from the electrode to the traps (Trap Assisted Tunneling, TAT) nearby the electrode. These electrons must be further excited to the conduction band again (Poole—Frenkel) leaving the ones in dee- per traps to hop to next trap sites to reach the anode when the insulator is thick enough [10].
Fig. 3. The decay time constant, t versus electric field intensity in HF SiN wit different thicknesses. Another contribution to the increase of the decay time constant with the electric filed intensity arises from the tunneled electrons from the electrode to the traps (Trap Assisted Tunneling, TAT) nearby the electrode. These electrons must be further excited to the conduction band again (Poole—Frenkel) leaving the ones in dee- per traps to hop to next trap sites to reach the anode when the insulator is thick enough [10].
Fig. 4. The stretch factor, # versus electric field intensity in HF SiN with different thicknesses. As can be seen from Figs. 3 and 5, the effect of the dielectric film thickness on the decay time constant is not clear when the same electric field intensity is applied to all film thicknesses. On the con- trary and according to our previous work, the decay time constant determined by Kelvin Probe Force Microscopy for both LF and HF
Fig. 4. The stretch factor, # versus electric field intensity in HF SiN with different thicknesses. As can be seen from Figs. 3 and 5, the effect of the dielectric film thickness on the decay time constant is not clear when the same electric field intensity is applied to all film thicknesses. On the con- trary and according to our previous work, the decay time constant determined by Kelvin Probe Force Microscopy for both LF and HF
Fig. 6. The decay time constant versus the dielectric film thickness for both LF and HF materials under +10 V. Fig. 5. Comparison between the decay time constants for both HF and LF SiN films with different thicknesses.
Fig. 6. The decay time constant versus the dielectric film thickness for both LF and HF materials under +10 V. Fig. 5. Comparison between the decay time constants for both HF and LF SiN films with different thicknesses.
Fig. 7. Stored charge versus the dielectric thickness under different electric filed intensities for: (a) LF and (b) HF SiN.
Fig. 7. Stored charge versus the dielectric thickness under different electric filed intensities for: (a) LF and (b) HF SiN.
Fig. 9. Stored charge, 9 versus the electric filed intensity for LF and HF SiN samples with 300 nm thickness. Fig. 9 presents the stored charge, oo calculated from Eq. (1), ver- sus the electric field intensity for both LF and HF SiN samples with 300 nm thickness. It is clear from the figure that the stored charge increases linearly with increasing the electric field intensity for both materials. This rule has been found to apply for all the inves- tigated dielectric thicknesses, LF and HF materials and under both bias polarities. On the other side, the dependence of stored charge on electric field intensity was found to follow a Poole-Frenkel like law as shown in Fig. 10. In principle, this rule has been found to ap- ply for both LF and HF materials for the whole thickness range and under both bias polarities. From the figure, the stored charge val- ues and the increase rates are not the same for LF and HF MIM capacitors with the same dielectric thickness. Basically, this can be related to the defect generation due to electron injection [13]. According to this the charge transport through traps in the band tails leads to a Poole-Frenkel like behavior. In such case the in- crease of trap concentration reduces the value of the Poole-Frenkel factor fpr of the classical Poole equation [13]. This is supported by the fact that the HF material exhibits a lower conductivity with re- spect to the LF one as well as the confinement of injected charges close to the metal contact. Here the authors would like to point out that the effect of the material properties constitutes an issue which is currently under further investigation.
Fig. 9. Stored charge, 9 versus the electric filed intensity for LF and HF SiN samples with 300 nm thickness. Fig. 9 presents the stored charge, oo calculated from Eq. (1), ver- sus the electric field intensity for both LF and HF SiN samples with 300 nm thickness. It is clear from the figure that the stored charge increases linearly with increasing the electric field intensity for both materials. This rule has been found to apply for all the inves- tigated dielectric thicknesses, LF and HF materials and under both bias polarities. On the other side, the dependence of stored charge on electric field intensity was found to follow a Poole-Frenkel like law as shown in Fig. 10. In principle, this rule has been found to ap- ply for both LF and HF materials for the whole thickness range and under both bias polarities. From the figure, the stored charge val- ues and the increase rates are not the same for LF and HF MIM capacitors with the same dielectric thickness. Basically, this can be related to the defect generation due to electron injection [13]. According to this the charge transport through traps in the band tails leads to a Poole-Frenkel like behavior. In such case the in- crease of trap concentration reduces the value of the Poole-Frenkel factor fpr of the classical Poole equation [13]. This is supported by the fact that the HF material exhibits a lower conductivity with re- spect to the LF one as well as the confinement of injected charges close to the metal contact. Here the authors would like to point out that the effect of the material properties constitutes an issue which is currently under further investigation.
Fig. 10. Pool-Frenkel law dependence of the stored charge for LF and HF samples.
Fig. 10. Pool-Frenkel law dependence of the stored charge for LF and HF samples.
Related topics:
MicroelectronicsRF MEMSSilicon NitrideElectrical and Electronic Engineering
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved