Next Article in Journal
The Relevant Approaches for Aligning Carbon Nanotubes
Next Article in Special Issue
Void Suppression in Glass Frit Bonding Via Three-Step Annealing Process
Previous Article in Journal
Electrical Performance of 28 nm-Node Varying Channel-Width nMOSFETs under DPN Process Treatments
Previous Article in Special Issue
Comprehensive Assessments in Bonding Energy of Plasma Assisted Si-SiO2 Direct Wafer Bonding after Low Temperature Rapid Thermal Annealing
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Combined Effect of TID Radiation and Electrical Stress on NMOSFETs

1
School of Electronics & Mechanical Engineering, Xidian University, Xi’an 710071, China
2
State Key Discipline Laboratory of Wide Bandgap Semiconductor Technology, School of Microelectronics, Xidian University, Xi’an 710071, China
3
Department of Electrical Engineering and Computer Science, Vanderbilt University, Nashville, TN 37235, USA
4
Beijing Microelectronics Technology Institute, Beijing 100076, China
*
Authors to whom correspondence should be addressed.
Micromachines 2022, 13(11), 1860; https://doi.org/10.3390/mi13111860
Submission received: 28 September 2022 / Revised: 17 October 2022 / Accepted: 27 October 2022 / Published: 29 October 2022
(This article belongs to the Special Issue Advanced Packaging for Microsystem Applications)

Abstract

:
The combined effect of total ionizing dose (TID) and electrical stress is investigated on NMOSFETs. For devices bearing both radiation and electrical stress, the threshold voltage shift is smaller than those only bearing electrical stress, indicating that the combined effect alleviates the degradation of the devices. The H bond is broken during the radiation process, which reduces the participation of H atoms in the later stage of electrical stress, thereby reducing the degradation caused by electrical stress. The positive charges of the oxide layer generated by radiation neutralize part of the tunneling electrons caused by electrical stress, and consume some of the electrons that react with the H bond, resulting in weaker degradation. In addition, the positive charges in shallow trench isolation (STI) generated by radiation create parasitic leakage paths at the interfaces of STI/Si, which increase the leakage current and reduce the positive shift of the threshold voltage. The parasitic effect generated by the positive charges of STI makes the threshold voltage of the narrow-channel device degrade more, and due to the gate edge effect, the threshold voltage of short-channel devices degrades more.

1. Introduction

With the development of semiconductor technology, the aerospace applications of integrated circuits have been greatly developed, and the extreme environment of space has seriously affected the operation of devices. Statistics show that the electronic circuits in satellites have more failures than mechanical failures, accounting for 54% of the total [1]. Unfavorable factors such as space radiation seriously affect the normal operation of the devices. Under space conditions, MOS devices will endure multiple radiation effects, e.g., single-event effects (SEEs) and total ionizing dose (TID) effects. In terms of total ionizing dose effects, much research has been conducted internationally, and it is generally believed that changes in parameters, such as the threshold voltage (Vth) of MOS devices after radiation, are mainly caused by the positive charges of the oxide layer and the interface traps at the SiO2/Si interface [2,3]. In addition, with the thinning of the gate oxide layer, the impact of shallow trench isolation (STI) becomes more significant [4,5,6]. The device reliability after radiation is also important for applications. Electrical stress on MOS devices causes a shift in threshold voltage, degradation in transconductance, etc. Changes in these device parameters may cause analog and digital circuit failures [7,8,9,10]. High-K gate dielectrics have been widely used in current technologies, which may lead to more severe degradations in NMOSFETs under positive gate bias [11]. Katsunori Onishi et al. believe that the NMOS degradation caused by positive gate stress was mainly due to the charges trapped in the HfO2 layer, rather than the interface trap [12,13]. N. Sa et al. proposed an improved version of the R-D model to explain the negative shift mechanism of threshold voltage during stress, which was attributed to the broken Si-O bond and the diffusion of O- in the HfO2 layer [14]. Mukhopadhyay et al. believed that the main reason for the degradation of high-K gate devices was the H-bond broken at the interface of HfO2/SiO2 and H- diffusion in the HfO2 layer due to the strong electric field [15]. Although there are many studies on the radiation or electrical stress of MOS devices, few studies have been reported on the combined effect of radiation and electrical stress on MOS devices. These combined effects may have new influences on the reliability of these devices. Therefore, we conducted in-depth research on the influence of the combined effects of radiation and electrical stress on device reliability. In this paper, we first analyze the combined effect of radiation and electrical stress on the characteristics of devices, discuss the effect of device structure size on the combined effect, and then determine the effects of parasitic gate and gate edge on the device.

2. Experiments

The experimental NMOSFETs samples were fabricated with a 28 nm process; the ratios of width and length were 1.5 μm/0.55 μm, 1.5 μm/1.1 μm, and 2.5 μm/0.55 μm, with EOT = 6.19 nm and operation voltage = 3.3 V. The selected devices were divided into two groups. The first group was for total ionizing dose radiation + electrical stress tests, and the second group was for electrical stress only. The radiation experiment was conducted at the Northwest Institute of Nuclear Technology. The radiation source was 60Co γ-rays and the dose rate was 61 rad (Si)/s. The total dose of 1000 krad(Si) was radiated to the NMOSFETs; the bias during radiation was ON state with gate voltage of Vg = 3.3 V, and Vs = Vd = Vsub = 0 V. Electrical stress experiments were performed after a total dose of 1000 krad(Si). During stress, gate voltage was 4.5 V (Vgstress = 4.5 V) and Vs = Vd = Vsub = 0 V. The electrical characteristics of the devices were tested at different time points. The total stress time was 1000 s, and the recovery time was 1000 s with Vg = Vs = Vd = Vsub = 0 V.

3. Experimental Results and Discussion

3.1. Initial I–V Characteristics of the Device

The initial characteristics of the devices were tested and the results are shown in Figure 1. The threshold voltage obtained from the transfer characteristics is Vth = 0.717 V, and the maximum transconductance is Gm-max = 3.63 × 10−5 S. The threshold voltage here is calculated using the constant current method [16], and the expression is shown below, where W and L are the gate width and gate length of NMOS devices, respectively. All the parameters are normal.
V t h = V g @ 0.1 × W L ( μ A )  

3.2. Combined Effect of Radiation and Electrical Stress

3.2.1. Effect of Radiation on I–V Characteristics

Figure 2 is a comparison of the transfer characteristics for NMOS devices before and after TID. The threshold voltage slightly shifted negatively and the subthreshold slope stretched out slightly after TID due to the positive charge generated in the oxide layer and the interface traps building up during irradiation [2,17]. In addition, device degradation is related to the HfO2 layer and shallow trench isolation (STI) [6,18]. The threshold voltage shift is composed of the oxide layer charges and the interface traps; the expression is [18]:
Δ V t h = Δ V o t + Δ V i t = t o x ε 0 ε o x q ( Δ N i t Δ N o t ) ,
where q is the electron charge, εox is the dielectric constant of the oxide, tox is the thickness of the oxide layer, ΔNot is the positive trap charge density of the oxide layer, and ΔNit is the interface trap charge density.
The calculation of interface traps is expressed by the following formula [19]:
Δ N i t   = 1 lg ( k T ) C o x q Δ S S   ,
where Cox is the oxide layer capacitance and ΔSS is the subthreshold swing variation.
The subthreshold swing (SS) is the change in gate voltage corresponding to each order of magnitude increase in subthreshold drain leakage current in the single logarithmic coordinate of the transfer curve. The calculation of SS is expressed by the following formula:
S S = d V g d ( log 10 I d ) ( V / dec )
where Vg and Id are from the transfer curve of the devices.
After calculation, the oxide layer charges and interface traps can be obtained, as shown in Figure 3.
It can be seen from Figure 3 that the oxide layer charges and interface traps increase rapidly before 200 krad (Si), and increase slowly after 200 krad (Si). The effect of oxide charges is greater than that of interface traps. For NMOSFETs, interface traps appear negatively charged and have a compensation effect on the positive charges generated by radiation. Therefore, even if the threshold voltage shifts caused by Not and Nit are large, the net shift of the NMOSFETs may be relatively small. As shown in Figure 3, the TID effect results in a relatively small negative shift of Vth after radiation by the compensation effect of ΔVot and ΔVit.

3.2.2. Combined Effect of Radiation and Electrical Stress on Long-Term Reliability

Figure 4 shows the variations in IdVg curves after electrical stress of TID and fresh devices. As can be seen from Figure 4 the devices all shift positively after electrical stress. The shift of the IdVg curve after TID + stress is less than that of the fresh + stress device, but the leakage current of the subthreshold is greater than that of fresh devices. The problem of increased leakage current is mainly caused by the oxide layer charge in STI. The effect of the positive charges in the STI cannot be ignored. Figure 5a shows the electric field at the STI. During the radiation process, the device not only generates positive charges in the gate oxide layer but also at the STI, generating an electric field at the STI, as shown in Figure 5a. Therefore, electrons can be attracted to the sides of the channel/STI and form two parasitic channels. The drain leakage current can be formed by electrons through the parasitic channels, as shown in Figure 5b. This makes the drain leakage current of the irradiated device increase greater than that of the non-irradiated device. The effect of STI is detailed in Section 3.3.
Under stress conditions, the threshold voltage is one of the most important degradation parameters. The threshold voltages are obtained from the transfer curves. It can be seen from Figure 6a that the threshold voltage degradation of the device which is applied a gate voltage stress after irradiation is smaller compared to that with a gate voltage applied directly. As can be seen in Figure 6b, in the double logarithmic coordinates, the threshold voltage shift shows a linear relationship to the stress time. In other words, the shift of the threshold voltage caused by the gate voltage stress conforms to the law of power function, similar to the effect of bias temperature instability, and its degradation expression is:
lg   ( Δ V t h ) = b + α lg   t ,
Thus,
Δ V t h = 10 b t α ,
where α is the slope of the linear fit in double logarithmic coordinates, and b is the curve intercept. In a non-irradiated device, the threshold voltage shifts by 0.2 V after 1000 s of stress; in an irradiated device, the threshold voltage shifts by 0.15 V after 1000 s of stress, which means that the radiation effect plays a role in weakening the degradation of the device during the biasing process. H will participate in both radiation and stress processes [20]. Unlike the reaction of H in the SiO2 layer of traditional MOSFETs [21], in high-K gate dielectrics, in addition to the Si/SiO2 interface, interface traps also occur at the SiO2/HfO2 interface. Figure 7 shows the mechanism of interface trap generation in a high-K dielectric.
When a positive gate bias stress is applied, the electrons in the channel inversion layer tunnel into the gate oxide layer, causing some Si-H or O-H bonds to break at the Si/SiO2 and SiO2/HfO2 interfaces, forming hydrogen atoms. Then, H atoms can react with X-H to form H2, thereby forming Si or O interface traps which appear negatively charged in NMOSFETs [22]. The model is as follows:
O-H + e = O + H; Si-H + e = Si + H,
X-H + H = X + H2
where X-H is Si-H or O-H.
Furthermore, during the fabrication process, a large number of oxygen vacancy defects in HfO2 will be generated, and they act as electron trapping centers [23] so that electrons can be trapped in the oxide layer to form negative oxide layer charges [24].
As analyzed above, the interface traps and negative oxide charges can cause the threshold voltage to shift positively [25], and they have a strengthening effect on the devices. The final positive shift of Vth is relatively large during the gate electrical stress.
It can be seen from Figure 8 that during the radiation process, the oxide layer will generate electron–hole pairs. Under the gate voltage, the electrons will move towards the HfO2 layer. Some of the electrons react with the Si-H bond at the SiO2/HfO2 interface and the HfO2 layer, consuming part of the H atoms. As a result, the H that can be reacted during the gate stress is reduced, and the degradation of the BTI-like effect is weakened. In addition, during the electrical stress experiment, a 4.5 V positive bias is applied to the gate, which will cause the electrons in the inversion layer to tunnel into the oxide layer. Since the radiation effect generates a large number of positive charges in the oxide layer, the tunneling electrons will neutralize part of the positive charges. This process consumes part of the electrons, thereby also weakening the BTI-like effect caused by gate stress. Based on the above two reasons, the degradation of devices with radiation combined with electrical stress is smaller than that of devices with only electrical stress, as shown in Figure 6.
Furthermore, the effects of the interface traps and oxide layer charges on the performance of the devices are significant [26]. With the varying of their densities, the threshold voltage shifts and the device reliability is affected. If the environmental condition such as radiation or electrical stress becomes more severe, the worse the device performance will be, and the more the reliability of the devices will degrade.
Figure 9 shows the changes in interface traps and oxide charges (ΔNit and ΔNot) with electrical stress time for fresh devices and devices after TID. Since the SiO2/HfO2 interface is inside the gate oxide layer, the defects generated by this interface can be regarded as oxide layer traps. In this way, the trap concentrations of ΔNit and ΔNot can be calculated according to Equations (2) and (3), as shown in Figure 9. It can be seen that ΔNot increases after stress for both devices, as the electrons are trapped in the oxide layer, forming oxidation layer charges. The ΔNit values of both devices are small, showing that the interface traps generated by the Si/SiO2 interface are small. The ΔNot of the device after TID + stress is smaller than that of the fresh one, as part of H has been consumed during the radiation process, and the tunneling electrons will neutralize part of the positive charges generated by the radiation effect. Therefore, the final ΔNot of the device after TID + stress is smaller.

3.3. Effect of Structure on Combined Effect of Radiation and Electrical Stress

3.3.1. Effect of Gate Width and STI Parasitic Gate

Figure 10 shows the relationship between the threshold voltage shift of different gate widths before and after radiation with time under electrical stress, where the stress time is 1000 s and the recovery time is 1000 s. It can be seen from Figure 10 that with the stress time accumulation, the NMOS devices show a certain degree of degradation, and the degradation trend gradually becomes saturated. The device will recover to a certain extent after 1000 s of recovery. This is because the gate stress is stopped and electrons stop tunneling, so the fracture process of the X-H bond stops, and the H ions diffuse back to the interface to passivate the X-H bond. As for the discussion before, the threshold voltage shift of the irradiated device is smaller than that of the non-irradiated device. Before and after radiation, the threshold voltage shifts are both larger during the stress process as the gate width becomes smaller, which is mainly related to the traps generated in STI by irradiation. In MOS devices, there are two kinds of oxide: gate oxide layer and shallow trench isolation (STI). With the thinning of the gate oxide, its influence on the device is reduced, and the influence of STI is highlighted [27]. In addition to the mechanical stress at the edge of the STI, there is also the electric field effect of the STI pointing to the channel due to the large number of positive charges generated in the STI region during radiation [28]. Therefore, the parasitic channel current of electrons is generated at both sides of STI and the parasitic gate effect is formed, making the threshold voltage shifts larger for devices with a narrower gate width.
In order to analyze the parasitic gate effect by STI, we simulated the physical characteristics of the STI side by SIVALCO. Figure 11 shows the three-dimensional simulation structure of the NMOS device. Figure 12a is a cross-sectional electric field diagram of the STI and the channel contact region along the gate width direction. As can be seen from Figure 12a, an electric field is generated at the interface between the STI and substrate silicon, pointing to the channel region from the STI region. Figure 12b shows the concentration of electrons generated at the substrate/STI interface due to the STI electric field, thereby forming a leakage channel. Then, the STI with positive charges by the TID effect can be seen as a parasitic gate. It can be seen from Figure 12a that as the gate width becomes smaller, the proportion 2a/W representing the influence of STI electric field regions increases. Therefore, under the same stress, the threshold voltage and other parameters of narrower devices degrade more.

3.3.2. Effect of Gate Length and Gate Edge Effect

Figure 13 shows the relationship between the threshold voltage shift of devices with different gate lengths before and after radiation as a function of time. It can be clearly seen that as the gate length decreases, the degradation of device parameters increases significantly. Figure 14 is a schematic diagram of damage to the device non-uniformly distributed along the channel. Because the gate edge is close to the source and drain doping and the electron concentration is high, it will cause excessive diffusion of the H atoms in the Si3N4 layer [29,30]. The manufacturing process will cause more potential damage at the gate edge [31], and the trap density in the gate edge region is higher than that in the center of the channel. As the channel length decreases, the ratio of the gate edge region to the total gate length increases, and the influence of the gate edge effect increases. Therefore, the dependence of the threshold voltage degradation of the device on the gate length is the same as that shown in Figure 13.

4. Conclusions

We studied the reliability of the combined effect of TID and electrical stress on NMOSFETs. First, the combined effect of radiation and electrical stress makes the degradation of threshold voltage and other characteristic parameters smaller, and the radiation effect weakens the ultimate degradation of the device. The electrons introduced by the radiation effect can destroy part of the H bonds, and the H bonds are an important reason for the degradation of the device during the stress process. Therefore, the degradation of the BTI-like effect is weakened. During the application of positive gate stress, the electrons tunnel into the oxide layer and neutralize part of the positive charges generated during the radiation process, which also reduces the degradation of the BTI-like effect. Secondly, the radiation effect will generate parasitic gates in the STI region, and the influence of the parasitic gates will increase as the gate width decreases. The gate edge region has more potential damage, resulting in a higher defect density than the middle region of the channel. The influence of the gate edge effect will increase as the gate length decreases. Therefore, the degradation of the device will be larger with a narrower and shorter gate.

Author Contributions

Funding acquisition, Y.C.; Methodology, M.W.; Project administration, Y.C.; Resources, Y.C., X.Z., E.Z., L.L. and L.W.; software, M.M., H.L. and Z.W.; Writing—original draft preparation, M.W. and Y.C; Writing—review and editing, M.W., Y.C., X.Z., E.Z., L.L., L.W., Y.W., W.T., X.M. and Y.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China (grant no. 62234013, 11690042, 12035019), the Cooperation Program of XDU-Chongqing IC Innovation Research Institute (grant no. CQIRI-2022CXY-Y10), National Major Scientific Research Instrument Projects (grant no. 61727804), and the Natural Science Foundation of Shaanxi Province (grant no. 2022-JM-386).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Zhang, S.; Shi, J.; Wang, J.L. Satellite on-board failure statistics and analysis. Spacecr. Eng. 2010, 19, 4l–46. [Google Scholar]
  2. Oldham, T.R.; Mclean, F.B. Total ionizing dose effects in MOS oxides and devices. IEEE. Trans. Nucl. Sci. 2003, 50, 483–499. [Google Scholar] [CrossRef] [Green Version]
  3. Schwank, J.R.; Shaneyfelt, M.R.; Fleetwood, D.M.; Felix, J.A.; Dodd, P.E.; Paillet, P.; Cavrois, V.F. Radiation effects in MOS oxides. IEEE. Trans. Nucl. Sci. 2008, 55, 1833–1853. [Google Scholar] [CrossRef]
  4. Faccio, F.; Cervelli, G. Radiation-induced edge effects in deep submicron CMOS transistors. IEEE. Trans. Nucl. Sci. 2005, 52, 2413–2420. [Google Scholar] [CrossRef] [Green Version]
  5. Zhou, H.; Liu, Y.; Zhang, Y. Total-ionizing-dose induced enhanced hot-carrier injection effect in the 130-nm partially depleted SOI I/O nMOSFETs. Jpn. J. Appl. Phys. 2020, 59, 031001. [Google Scholar] [CrossRef]
  6. Fleetwood, D.M. Total-Ionizing-Dose effects, border traps, and 1/f noise in emerging MOS technologies. IEEE. Trans. Nucl. Sci. 2020, 67, 1216–1240. [Google Scholar] [CrossRef]
  7. Schroder, D.K.; Babcock, J.A. Negative bias temperature instability: Road to cross in deep submicron silicon semiconductor manufacturing. J. Appl. Phys. 2003, 94, 1–18. [Google Scholar] [CrossRef] [Green Version]
  8. Ullmann, B.; Jech, M.; Tyaginov, S.; Waltl, M.; Lllarionov, Y.Y.; Grill, A.; Puschkarsky, K.; Reisinger, H.; Grasser, T. The impact of mixed negative bias temperature instability and hot carrier stress on single oxide defects. IEEE. Inter. Reliab. Phys. Symp. 2017, XT-10.1–XT-10.6. [Google Scholar]
  9. Jech, M.; Ullmann, B.; Rzepa, G.; Tyaginov, S.; Grill, A.; Waltl, M.; Jabs, D.; Jungemann, C.; Grasser, T. Impact of mixed negative bias temperature instability and hot carrier stress on MOSFET characteristics—Part II: Theory. IEEE. Trans. Elec. Dev. 2019, 66, 241–248. [Google Scholar] [CrossRef]
  10. Zhou, X.J.; Fleetwood, D.M.; Felix, J.A.; Gusev, E.P.; Emic, D. Bias-temperature instabilities and radiation effects in MOS devices. IEEE. Trans. Nucl. Sci. 2006, 52, 2231–2238. [Google Scholar] [CrossRef]
  11. Zafar, S.; Kim, Y.; Narayanan, V.; Cabral, C.; Paruchuri, V.; Doris, B.; Stathis, J.; Callegari, A.; Chudzik, M. A Comparative Study of NBTI and PBTI (Charge Trapping) in SiO2/HfO2 Stacks with FUSI, TiN, Re Gates. Symp. VLSI. Technol. Dig. Tech. Pap. 2006, 23–25. [Google Scholar]
  12. Waldhoer, D.; El-Sayed, A.M.B.; Wimmer, Y.; Waltl, M.; Grasser, T. Atomistic Modeling of Oxide Defects. Noise. Nanosc. Semicon. Dev. 2020, 609–648. [Google Scholar]
  13. Onishi, K.; Choi, R.; Kang, S.K.; Cho, H.J.; Kim, Y.H.; Nieh, R.E.; Han, J.; Krishnan, S.A.; Akbar, M.S.; Lee, J.C. Bias-temperature instabilities of polysilicon gate HfO/sub2/ MOSFETs. IEEE. Trans. Nucl. Sci. 2003, 50, 1517–1524. [Google Scholar] [CrossRef]
  14. Sa, N.; Kang, J.F.; Yang, H.; Liu, X.Y.; He, Y.D.; Han, R.Q.; Ren, C.; Yu, H.Y.; Chan, D.S.H.; Kwong, D.L. Mechanism of positive-bias temperature instability in sub-1-nm TaN/HfN/HfO2 gate stack with low preexisting traps. IEEE Elec. Dev. Lett. 2005, 26, 610–612. [Google Scholar] [CrossRef]
  15. Mukhopadhyay, S.; Parihar, N.; Goel, N.; Mahapatra, S. A Comprehensive DC and AC PBTI Modeling Framework for HKMG n-MOSFETs. IEEE Trans. Elec. Dev. 2017, 64, 1474–1481. [Google Scholar] [CrossRef]
  16. Ortiz-Conde, A.; Sánchez, F.J.G.; Liou, J.J.; Cerdeira, A.; Estrada, M.; Yue, Y. A review of recent MOSFET threshold voltage extraction methods. Microelectron. Reliab. 2002, 42, 583–596. [Google Scholar]
  17. Hughes, H.L.; Benedetto, J.M. Radiation effects and hardening of MOS technology: Devices and circuits. IEEE. Trans. Nucl. Sci. 2003, 50, 500–521. [Google Scholar]
  18. Li, N.L.; Yu, Q.; Wang, K.; Yu, W.H.; Dong, Z.X. Study on simulation models for γ total dose radiation effects in MOSFET structure. Microelectronics 2013, 43, 445–448. [Google Scholar]
  19. Franco, J.; Kaczer, B.; Mukhopadhyay, S.; Duban, P.; Weckx, P.; Roussel, P.J.; Chiarella, T.; Ragnarsson, L.A.; Troiman, L.; Horiguchi, N.; et al. Statistical model of the NBTI-induced threshold voltage, subthreshold swing, and transconductance degradations in advanced p-FinFETs. IEEE Int. Elec. Dev. Meet. 2016, 15.3.1–15.3.4. [Google Scholar]
  20. Mukhopadhyay, S.; Joshi, K.; Chaudhary, V.; Goel, N.; De, S.; Pandey, R.; Murali, K.; Mahapatra, S. Trap Generation in IL and HK layers during BTI / TDDB stress in scaled HKMG N and P MOSFETs. IEEE Int. Reliab. Phys. Symp. 2014, GD.3.1–GD.3.11. [Google Scholar]
  21. Griscom, D.L. Hydrogen model for radiation-induced interface states in SiO2-on-Si Structures: A review of the evidence. J. Electron. Mater. 1992, 21, 763–767. [Google Scholar] [CrossRef]
  22. Parihar, N.; Goel, N.; Chaudhary, A.; Mahapatra, S. A Modeling Framework for NBTI Degradation Under Dynamic Voltage and Frequency Scaling. IEEE Trans. Elec. Dev. 2016, 63, 946–953. [Google Scholar] [CrossRef]
  23. Robertson, J.; Xiong, K.; Tse, K.Y. Importance of Oxygen Vacancies in High K Gate Dielectrics. IEEE Inter. Confer. Integr. Circ. Des. Tech. 2007, 1–4. [Google Scholar]
  24. Kerber, A.; Cartier, E.; Pantisano, L.; Degraeve, R.; Kauerauf, T.; Kim, Y.; Hou, A.; Groeseneken, G.; Maes, H.E.; Schwalke, U. Origin of the threshold voltage instability in SiO2/HfO2 dual layer gate dielectrics. IEEE Elec. Dev. Lett. 2003, 24, 87–89. [Google Scholar] [CrossRef]
  25. Abliz, A.; Xue, X.X.; Liu, X.Q.; Li, G.L.; Tang, L.M. Rational design of hydrogen and nitrogen co-doped ZnO for high performance thin-film transistors. Appl. Phys. Lett. 2021, 118, 123504. [Google Scholar] [CrossRef]
  26. Abliz, A.; Xu, L.; Wan, D.; Duan, H.M.; Wang, J.L.; Wang, C.L.; Luo, S.J. Effects of yttrium doping on the electrical performances and stability of ZnO thin-film transistors. Appl. Surf. Sci. 2019, 475, 565–570. [Google Scholar] [CrossRef]
  27. Chung, S.S.; Yeh, C.H.; Feng, H.J.; Lai, C.S.; Yang, J.J.; Chen, C.C.; Jin, Y.; Chen, S.C.; Liang, M.S. Impact of STI on the reliability of narrow-width pMOSFETs with advanced ALD N/O gate stack. IEEE Trans. Dev. Mater. Reliab. 2006, 6, 95–101. [Google Scholar] [CrossRef]
  28. Cui, J.W.; Zheng, Q.W.; Yu, D.Z.; Zhou, H.; Su, D.D.; Ma, T.; Wei, Y.; Yu, X.F.; Guo, Q. Effect of channel width on NBTI in 65nm PMOSFET. Tien. Tzu. Hsueh. Pao. Acta. Elec. Sinica. 2018, 46, 1128–1132. [Google Scholar]
  29. Jin, L.; Xu, M.Z. Effect of channel length on NBTI in sub-100nm CMOS technology. IEEE Int. Nanoelectron. Conf. 2008, 597–600. [Google Scholar]
  30. Lu, C.Y.; Lin, H.C.; Lee, Y.J. Dynamic NBTI characteristics of PMOSFETs with PE-SiN capping. Microelectron. Reliab. 2007, 47, 924–929. [Google Scholar] [CrossRef]
  31. Cao, Y.R.; Ma, X.H.; Hao, Y.; Tian, W.C. Effect of channel length and width on NBTI in ultradeep sub-micron PMOSFETs. Chinese. Phys. Lett. 2010, 27, 037301. [Google Scholar]
Figure 1. I−V characteristics for a typical NMOSFET in this work: (a) output characteristics and (b) transfer and transconductance curves.
Figure 1. I−V characteristics for a typical NMOSFET in this work: (a) output characteristics and (b) transfer and transconductance curves.
Micromachines 13 01860 g001
Figure 2. Drain current as a function of gate voltage for devices before and after TID radiation. The W/L of the devices is 1.5 µm/0.55 µm.
Figure 2. Drain current as a function of gate voltage for devices before and after TID radiation. The W/L of the devices is 1.5 µm/0.55 µm.
Micromachines 13 01860 g002
Figure 3. ΔNot and ΔNit at different doses during radiation.
Figure 3. ΔNot and ΔNit at different doses during radiation.
Micromachines 13 01860 g003
Figure 4. Drain current variations in IdVg curves after electrical stress of (a) TID and (b) fresh devices. The W/L of the devices is 1.5 µm/0.55 µm.
Figure 4. Drain current variations in IdVg curves after electrical stress of (a) TID and (b) fresh devices. The W/L of the devices is 1.5 µm/0.55 µm.
Micromachines 13 01860 g004
Figure 5. (a) Schematic diagram of STI electric field. (b) Schematic diagram of the parasitic channels generated by radiation in STI.
Figure 5. (a) Schematic diagram of STI electric field. (b) Schematic diagram of the parasitic channels generated by radiation in STI.
Micromachines 13 01860 g005
Figure 6. Degradation comparison of threshold voltage under electrical stress only and TID electrical stress: (a) linear coordinates; (b) double logarithm coordinates.
Figure 6. Degradation comparison of threshold voltage under electrical stress only and TID electrical stress: (a) linear coordinates; (b) double logarithm coordinates.
Micromachines 13 01860 g006
Figure 7. Mechanism of interface trap generation in a high-K dielectric.
Figure 7. Mechanism of interface trap generation in a high-K dielectric.
Micromachines 13 01860 g007
Figure 8. Radiation produces defects in the oxide layer.
Figure 8. Radiation produces defects in the oxide layer.
Micromachines 13 01860 g008
Figure 9. The relationship between ΔNit and ΔNot and electrical stress time for the fresh devices and devices after TID.
Figure 9. The relationship between ΔNit and ΔNot and electrical stress time for the fresh devices and devices after TID.
Micromachines 13 01860 g009
Figure 10. The relationship between threshold voltage shift with different gate widths as a function of time.
Figure 10. The relationship between threshold voltage shift with different gate widths as a function of time.
Micromachines 13 01860 g010
Figure 11. MOSFET simulation structure.
Figure 11. MOSFET simulation structure.
Micromachines 13 01860 g011
Figure 12. Device simulation results (STI in the gate width direction): (a) electric field and (b) electron density of the STI side.
Figure 12. Device simulation results (STI in the gate width direction): (a) electric field and (b) electron density of the STI side.
Micromachines 13 01860 g012
Figure 13. The relationship of threshold voltage shifts of devices with different gate lengths as a function of time.
Figure 13. The relationship of threshold voltage shifts of devices with different gate lengths as a function of time.
Micromachines 13 01860 g013
Figure 14. Schematic diagram of damage to the device non-uniformly distributed along the channel.
Figure 14. Schematic diagram of damage to the device non-uniformly distributed along the channel.
Micromachines 13 01860 g014
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Cao, Y.; Wang, M.; Zheng, X.; Zhang, E.; Lv, L.; Wang, L.; Ma, M.; Lv, H.; Wang, Z.; Wang, Y.; et al. Combined Effect of TID Radiation and Electrical Stress on NMOSFETs. Micromachines 2022, 13, 1860. https://doi.org/10.3390/mi13111860

AMA Style

Cao Y, Wang M, Zheng X, Zhang E, Lv L, Wang L, Ma M, Lv H, Wang Z, Wang Y, et al. Combined Effect of TID Radiation and Electrical Stress on NMOSFETs. Micromachines. 2022; 13(11):1860. https://doi.org/10.3390/mi13111860

Chicago/Turabian Style

Cao, Yanrong, Min Wang, Xuefeng Zheng, Enxia Zhang, Ling Lv, Liang Wang, Maodan Ma, Hanghang Lv, Zhiheng Wang, Yongkun Wang, and et al. 2022. "Combined Effect of TID Radiation and Electrical Stress on NMOSFETs" Micromachines 13, no. 11: 1860. https://doi.org/10.3390/mi13111860

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop