Next Article in Journal
Compressed-Sensing-Based Time–Frequency Representation for Disturbance Characterization of Maglev On-Board Distribution Systems
Next Article in Special Issue
Analysis of State-of-the-Art Spin-Transfer-Torque Nonvolatile Flip-Flops Considering Restore Yield in the Near/Sub-Threshold Voltage Region
Previous Article in Journal
Synchronization and Sampling Time Analysis of Feedback Loop for FPGA-Based PMSM Drive System
Previous Article in Special Issue
Robust Offset-Cancellation Sense Amplifier for an Offset-Canceling Dual-Stage Sensing Circuit in Resistive Nonvolatile Memories
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Partial Isolation Type Buried Channel Array Transistor (Pi-BCAT) for a Sub-20 nm DRAM Cell Transistor

Department of ICT Convergence System Engineering, Chonnam National University, Gwangju 61186, Korea
*
Authors to whom correspondence should be addressed.
Electronics 2020, 9(11), 1908; https://doi.org/10.3390/electronics9111908
Submission received: 14 October 2020 / Revised: 4 November 2020 / Accepted: 5 November 2020 / Published: 13 November 2020

Abstract

:
In this paper, we propose a new buried channel array transistor structure to solve the problem of current leakage occurring in the capacitors of dynamic random-access memory (DRAM) cells. This structure has a superior off current performance compared with three previous types of structures. In particular, the proposed buried channel array transistor has a 43% lower off current than the conventional asymmetric doping structure. Here, we show the range of the effective buried insulator parameter according to the depth of the buried gate, and we effectively show the range of improvement for the off current.

1. Introduction

As the dynamic random-access memory (DRAM) cell size decreases, DRAM reduces the size of the line width and length of the gate [1,2]. As a result, this increases the short-channel effect of the conventional planar MOSFET [3,4]. In order to increase the channel length of the gate, a recessed channel array transistor that increases the length of the channel by providing a recessed channel area below the existing gate is proposed [5]. The recessed channel array transistor improves the short channel effect, but has the problem that the long channel length decreases the on current [6]. Likewise, Fin field effect transistor(FinFET) with three-sided channels improves the short-channel effect and on current, but the off current increases because of the increase in the area of the overlapped region of the gate and drain [7,8]. To compensate for this, a saddle-type FinFET is proposed by adding gates on both sides of the recessed channel array transistor [9,10,11]. However, the saddle-type FinFET has an increased gate induced drain leakage (GIDL) because of its wider overlapped regions of gate and drain compared with the recessed channel array transistors [12]. RFinFET is a structure that improves gate induced drain leakage, and unlike a saddle-type FinFET structure, the side gate areas of RFinFET are installed only in the channel area [13,14]. The buried channel array transistor that is currently being manufactured is a structure in which the gate is buried in the structure of RFinFET, and the overlapped region between the gate and drain is reduced [15,16]. However, there is a limit to the amount the overlapped region of gate and drain under the structure can be reduced. In addition, the continuous reduction in the size of the line width requires a cell size of less than 20 nm, and the high doping concentration increases the GIDL [17]. To solve this problem, the existing structure currently being manufactured needs to be improved. This paper proposes a structure that is partially isolated in the area below the storage node of the buried channel array transistor (Pi-BCAT) for a DRAM cell transistor of less than 20 nm [18,19,20,21]. In addition, the current of buried channel-array transistors with the same size and similar shape as the existing fabricated structure (asymmetric doping structure) are compared [22,23,24]. Next, the parameters of the buried insulator are introduced, and the optimized values are proposed. The structures presented in this paper are fabricated using a 3D device simulation (sentaurus TCAD). These device structures use a Gaussian distribution of the doping concentration in order to provide a reliable doping profile [25,26,27].

2. Device Structure

The partial isolation type buried channel array transistor (Pi-BCAT) is a structure that has a buried insulator at a certain depth from the storage node of the buried channel array transistor (BCAT). Figure 1 shows the 3D structure of a Pi-BCAT. In Figure 1, the first of the important parameters, Lin, is the distance between the gate oxide and the insulator, while the second is the silicon film thickness, which is the distance from the storage node contact surface to the buried insulator.
The third is the gate depth, which means the distance to the buried gate based on the storage node contact surface. Figure 2 shows a cross-sectional view of x–z when y = 0 for two saddle RFinFET structures and three buried channel array transistor structures all possessing the same cell sizes. Figure 2a shows the structure of a saddle RFinFET, while Figure 2b shows a Pi-RFinFET with a buried insulator in the doping region at the bottom of the storage node in the structure of Figure 2a [27]. Figure 2c shows an asymmetric doping structure (asymmetry) with a short doping junction depth at the bottom of the storage node, and a long doping junction depth at the bottom of the bit line contact. Figure 2d shows a doping junction depth with a longer area at the bottom of the storage node than in Figure 2c. Figure 2e shows a Pi-BCAT structure with a buried insulator in the doping region below the storage node in the structure of Figure 2d. The device parameters in Figure 2 are as follows: In Figure 2a,b, gate recess depth = 120 nm and the peak doping concentration is 1.05 × 1020 cm−3. In Figure 2c–e, gate depth = 80 nm, gate recess depth = 150 nm, peak doping concentration of BCAT = 1.45 × 1020 cm−3, short junction depth = 82–92 nm, long doping junction depth = 134–141 nm. In Figure 2b, the parameters of the insulator are Lin = 6 nm, silicon film thickness = 30 nm, end depth = 95 nm, and in Figure 2e, Lin = 6 nm, silicon film thickness = 60 nm, and end depth = 145 nm.

3. Results and Discussion

Figure 3 numerically compares the on/off current of five structures (Saddle RFinFET, Pi-RFinFET, asymmetry, long doping junction BCAT, and Pi-BCAT). We defined the gap between the equipotential lines as the potential drop width (PDW). Compared with the buried channel array transistor with an asymmetric doping structure (Figure 2c), the saddle RFinFET has a lower peak doping concentration than the other three buried channel array transistor structures, because the saddle RFinFET has a wider overlapped region between the gate and the drain than the BCAT structure. The data to note in Figure 3 is for the off current, and the off current of each device is Ioff_saddle = 7.7 × 10−13 A, Ioff_Pi-RFinFET = 4.89 × 10−13 A, Ioff_Asymmetry_BCAT = 5.95 × 10−14 A, Ioff_Long_doping_junction = 1.84 × 10−13 A, and Ioff_Pi-BCAT = 2.54 × 10−14 A. Figure 3 shows that the off current is higher than that of other BCAT structures, even though the peak doping concentration is lowered. Pi-RFinFET is a structure manufactured according to the reference, and an insulator is added to the doping region at the bottom of the storage node of the saddle RFinFET. As a result, the on current decreases because of the buried insulator, but the off current decreases by increasing the potential drop width of the overlapped region between the gate and the drain. In the case of a long doping junction structure, the doping region is relatively wider than that of the asymmetric doping structure. As a result, the channel length is short and the on current is high. However, the off current of the long doping junction structure increases relatively compared to the asymmetry doping structure, because the overlapped region between the gate and the drain increases. In Pi-BCAT, it partially offsets the on current reduction phenomenon that occurs in the buried insulator as it increases the on current, which is an advantage of the long doping junction structure. In addition, as the dielectric constant of the insulator (SiO2: 3.9) is smaller than that of silicon (Si: 11.8), the internal electric field of the insulator is small. In addition, by increasing the potential drop width in the silicon region around the insulator according to the boundary conditions of Gauss’s law, GIDL becomes lower. As a result, the on current decreases by 20% compared with the asymmetry doping structure, but the off current decreases by 58%.
Figure 4 shows a linear Log current graph of five structures (saddle RFinFET, Pi-RFinFET, asymmetry, long doping junction BCAT, and Pi-BCAT). The Ioff values of each structure in the red circle corresponding to the gate voltage = −0.5 V are in the following order: Pi-BCAT < asymmetry < long doping junction BCAT < Pi-RFinFET < saddle RFinFET. The structures of saddle RFinFET and BCAT have different values of Ioff because of the difference in the area of the gate, while asymmetry and long doping junction BCAT have different Ioff, owing to the difference in the overlapped region of the gate and drain. In the case of the Pi-RFinFET and Pi-BCAT structures, in which the insulators are buried, they have a smaller Ioff than the general structures (saddle RFinFET and long doping junction BCAT) for the difference in permittivity and the boundary conditions of Gauss’s law between the insulator and silicon.
Figure 5 shows a numerical comparison of the gm.max (maximum transconductance) and subthreshold slope (SS) of five structures (saddle RFinFET, Pi-RFinFET, asymmetry BCAT, long doping junction BCAT, and Pi-BCAT). In the case of gm.max, compared with the asymmetric doping structure (asymmetry), the long doping junction structure has a wide doping region and a short channel length, therefore gm.max is high. In Pi-BCAT, compared with the asymmetric doping structure, because of the buried insulator, the doping area decreases and the resistance increases, so gm.max decreases. In the case of SS, the difference between N-type doping and P-type doping is largely designed for the long doping junction structure compared with the asymmetric doping structure. As a result, the depletion width increases, and the SS of the long doping junction structure is relatively large compared with the asymmetry structure. Compared with the long doping junction BCAT, Pi-BCAT has a smaller electric field. As a result, SS decreases because the depletion width becomes smaller.
Figure 6 is a numerical comparison of the Vth and drain induced barrier lowering(DIBL) of the five structures (saddle RFinFET, Pi-RFinFET, asymmetry BCAT, long doping junction BCAT, and Pi-BCAT). In the case of Vth, as the long doping junction structure has a higher N-type doping concentration compared with the region than the asymmetric doping structure (asymmetry), the energy level of the N-type doping region is relatively low. As a result, Vth is relatively low. In Pi-BCAT, the energy level of the insulator (SiO2) is higher than that of silicon. Because of the relatively high energy level of the insulator (SiO2), the energy level of silicon is raised by the boundary conditions of Gauss’s law, and Vth is raised. In the case of DIBL, the gate channel length of the long doping junction BCAT is shorter than the gate channel length of the asymmetric doping structure, so the DIBL becomes larger. In the case of Pi-BCAT, the DIBL is small, as the insulator causes most of the high potential to be located in the doping region at the bottom of the storage nodes.
Figure 7 compares the Ion and Ioff values of Pi-BCAT according to the length of Lin in the Pi-BCAT. As Lin decreases, both the on and off currents decrease, showing that the off-current’s decrease is relatively significantly compared with that of the on current. Comparing the on and the off currents of the asymmetric doping structure as a measure, the on current is 80% and off current is 42% at Lin = 6 nm for Pi-BCAT.
Figure 8 shows the electrostatic potential distribution according to the length of the Lin in Pi-BCAT (VGS = −0.5 V and VDS= 1.2 V), with Figure 8a Lin = 6 nm and Figure 8b Lin = 12 nm. The red dotted circle area is the high potential area where band to band tunneling(BTBT) occurs. Dotted arrows indicate a potential drop width from 1.2 V to 0 V. As the length of Lin increases, the potential in the BTBT region where the gate and drain overlap increases. In addition, as the buried insulator area decreases, the silicon area increases. Therefore, the potential drop width near the gate oxide decreases. As a result, the BTBT region has a high potential and a strong electric field, so the off current increases.
Figure 9 shows the electric field and BTBT in the cut line. The BTBT regions of Lin = 6 nm and 12 nm are marked with blue lines. At Lin = 6 nm and 12 nm, the value of the peak electric field is almost the same. When Lin = 6 nm, the BTBT region is 68 to 86 nm, and when Lin = 12 nm, the BTBT region is 72 nm to 94 nm. In the case of Lin= 6 nm in the BTBT region, the integral of the E-Field is 1.29 × 10−2 V/μm, and in Lin = 12 nm, the integral of the total electric field is 1.51 × 10−2 V/μm. When Lin is long, the area of the insulator decreases and the area of the silicon increases, so the region of the BTBT widens and the electric field increases because the influence of the insulator decreases. This increases the off current.
Figure 10 compares the on/off current values using the silicon film thickness of Pi-BCAT. Increasing the silicon film thickness means that the high doping area becomes larger, and the on current increases. Likewise, until the silicon film thickness is 60 nm, the off current does not show a big change, but when the silicon film thickness is more than 60 nm, it increases sharply. This is because when the silicon film thickness of the buried insulator increases, the potential drop width in the area where BTBT mainly occurs decreases, so GIDL increases. However, when the silicon film thickness exceeds the gate depth, the influence on the potential drop width of the area where BTBT mainly occurs decreases and reaches saturation.
Figure 11 shows the potential distribution plot according to the silicon film thickness in Pi-BCAT (VGS= −0.5 V and VDS = 1.2 V), with (a) silicon film thickness = 30 nm, (b) silicon film thickness = 60 nm, and (c) silicon film thickness = 90 nm. When the silicon film thickness is small, the Pi-BCAT has a long potential drop width because of the boundary conditions of Gauss’s law of the insulator. However, as the size of the silicon film thickness gets closer to the gate depth, the potential drop width becomes smaller, and when the silicon film thickness is 70 nm compared with a gate depth = 80 nm, the potential drop width becomes shorter than the existing Pi-BCAT. When the silicon film thickness becomes longer than 70 nm, the position of the insulator gradually deviates from the area where BTBT mainly occurs, and as a result, after decreasing, GIDL becomes constant.
Figure 12 shows the on/off current graph of Pi-BCAT according to the length of Lin in Pi-BCAT with a gate depth of 70 nm. Similar to Figure 6, the on/off currents decrease as Lin decreases, but compared to the on current, the off current decreases significantly. In particular, it corresponds to 80% of the on current of the asymmetry junction BCAT at Lin = 5 nm, because the doping concentration is relatively concentrated in the contact area compared with a gate depth = 80 nm.
Figure 13 plots the on/off current according to the silicon film thickness of Pi-BCAT when the gate depth is 70 nm. Similar to Figure 9, as the silicon film thickness increases, both the on/off currents increase, but compared with the on current, the off current increases significantly above the silicon film thickness = 50 nm. However, when the silicon film thickness exceeds the gate depth, the effect on the potential drop width in the area where BTBT mainly occurs is reduced, and the off current reaches a saturation value and then decreases.
Figure 14 shows the on/off current graph of Pi-BCAT according to the length of the Lin in Pi-BCAT with a gate depth = 90 nm. Similar to Figure 6 and Figure 11, as Lin decreases, the on/off currents decrease, but the off current decreases significantly compared with the on current. In particular, it corresponds to 80% of the on current of the asymmetry doping structure at Lin = 6 nm, and the off current tends to decrease significantly.
Figure 15 shows the on/off current graph according to the silicon film thickness of Pi-BCAT at 90 nm of the gate depth. Similar to Figure 9 and Figure 12, as the silicon film thickness increases, both the on/off currents increase. When the silicon film thickness exceeds 70 nm, the off current increases sharply. However, when the silicon film thickness exceeds the gate depth, the off current reaches the saturation value, and then decreases.
Figure 16 shows the optimal parameter at each gate depth of 70, 80, and 90 nm in the Pi-BCAT. According to the gate depth, the optimal value of Lin gradually increases. This is because as the gate depth increases, the doping concentration near the gate decreases. As a result, the optimal Lin increases. The BTBT generation region is determined according to the gate depth. Therefore, the longer the gate depth, the longer the optimal silicon film thickness.
Table 1 shows the Lin and silicon film thickness of Pi-BCAT according to 80% of the on current in asymmetry BCAT at gate depths = 70 nm, 80 nm, and 90 nm. In addition, the off current of Pi-BCAT compared with asymmetry BCAT according to each gate depth of 70 nm, 80 nm, and 90 nm are 38%, 33%, and 38%. This is a three times improvement in the off current performance of asymmetry BCAT.

4. Conclusions

This paper proposes a structure with a partial isolation area under the storage node of the buried channel array transistor. In the partial insulation buried channel array transistor, as the length of Lin for the buried insulator decreases, the on current decreases, but the off current falls significantly. In addition, the optimum point of length of the silicon film thickness of the buried insulator is determined by the gate depth. Therefore, the degree of the buried insulator can be set through the on/off current trend data according to the Lin and silicon film thickness provided in this paper. In particular, although the partial isolation type buried channel array transistor structure with an optimized buried insulator compared with the conventional structure with an asymmetry doping structure shows some loss of the on current, the off current has the relatively very low value of 40%, showing an improved performance in terms of reducing leakage current in memory devices.

Author Contributions

Formal analysis, J.-h.P.; Investigation, G.K.; Supervision, M.J.L.; Writing—original draft, J.-s.L.; Writing—review & editing, H.D.C. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported in part by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT; grant number 2018R1A2B600821613) and in part by Korea Institute for Advancement of Technology(KIAT) grant funded by the Korea Government(MOTIE) (P0011931, The Establishment Project of Industry-University Fusion District) and in part by Samsung Electronics (grant number 2019-0184).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Kim, K. Future Silicon Technology. In Proceedings of the 2012 Proceedings of the ESSCIRC(ESSCIRC), Bordeaux, France, 17–21 September 2012; pp. 1–6. [Google Scholar]
  2. Hong, S. Memory Technology Trend and Future Challenges. In Proceedings of the 2010 International Electron Devices Meeting, San Francisco, CA, USA, 6–8 December 2010; pp. 1–4. [Google Scholar]
  3. Duvvury, C. A guide to short-channel effects in MOSFETs. IEEE Circuits Devices Mag. 1986, 2, 6–10. [Google Scholar] [CrossRef]
  4. Chaudhry, A.; Jagadesh, M.K. Controlling short-channel effects in deep-submicron SOI MOSFETs for improved reliability: A review. IEEE Trans. Device Mater. Reliab. 2004, 4, 99–109. [Google Scholar] [CrossRef] [Green Version]
  5. Kim, J.Y.; Lee, C.S.; Kim, S.E.; Chung, I.B.; Choi, Y.M.; Park, B.J.; Lee, J.W.; Kim, D.I.; Hwang, Y.S.; Hwang, D.S. The Breakthrough in data retention time of DRAM using Recess-Channel-Array Transistor(RCAT) for 88nm feature size and beyond. In Proceedings of the 2003 Symposium on VLSI Technology, Kyoto, Japan, 10–12 June 2003; pp. 11–12. [Google Scholar]
  6. Kim, L.J.Y.; Oh, H.J.; Lee, D.S.; Kim, D.H.; Kim, S.E.; Ha, G.W.; Kim, H.J.; Kang, N.J.; Park, J.M.; Hwang, Y.S. S-RCAT (sphere-shaped-recess channel-array transistor) technology for 70 nm DRAM feature size and beyond. In Proceedings of the VLSI Symposium, Kyoto, Japan, 14–16 June 2005; pp. 34–35. [Google Scholar]
  7. Kondo, M.; Katsumata, R.; Aochi, H.; Hamamoto, T.; Ito, S.; Aoki, N.; Wada, T. A FinFET design based on three-dimensional process and device simulations. In Proceedings of the International Conference on Simulation of Semiconductor Processes and Devices, Boston, MA, USA, 3–5 September 2003; pp. 179–182. [Google Scholar]
  8. Mahmoud, S.B.; Issa, H.H.; Saleh, M.E.; Ragai, H.F. Low Leakage Current Symmetrical Dual-k 7 nm Trigate Bulk Underlap FinFET for Ultra Low Power Applications. IEEE Access 2019, 7, 17256–17262. [Google Scholar]
  9. Chung, S.W.; Lee, S.D.; Jang, S.A.; Yoo, M.S.; Kim, K.O.; Chung, C.O.; Cho, S.Y.; Cho, H.J.; Lee, L.H.; Hwang, S.H. Highly scalable saddle-fin (S-Fin) transistor for sub 50 nm DRAM technology. In Proceedings of the VLSI Symposium, Honololu, HI, USA, 13–15 June 2006; pp. 147–148. [Google Scholar]
  10. Poljak, M.; Jovanovic, V.; Suligoj, T. Improving bulk FinFET DC performance in comparison to SOI FinFET. Microelectron. Eng. 2009, 86, 2078–2085. [Google Scholar] [CrossRef]
  11. Crupi, G.; Schreurs, D.; Raskin, J.P.; Caddemi, A. A comprehensive review on microwave FinFET modeling for progressing beyond the state of art. Solid State Electron. 2013, 80, 81–95. [Google Scholar] [CrossRef]
  12. Kerber, P.; Zhang, Q.; Koswatta, S.; Bryant, A. GIDL in Doped and Undoped FinFET Devices for Low-Leakage Applications. IEEE Electron Device Lett. 2013, 34, 6–8. [Google Scholar] [CrossRef]
  13. Lee, M.J.; Jin, S.H.; Baek, C.K.; Hong, S.M.; Park, S.Y.; Park, H.H.; Lee, S.D.; Chung, S.W.; Jeong, J.G.; Hong, S.G. A proposal on an optimized device structure with experimental studies on recent devices for the DRAM cell transistor. IEEE Trans. Electron Devices 2007, 54, 3325–3335. [Google Scholar] [CrossRef]
  14. Ryu, S.W.; Min, K.; Shin, J.; Kwon, H.; Nam, D.; Oh, T.; Jang, T.S.; Yoo, M.; Kim, Y.; Hong, S. Overcoming the reliability limitation in the ultimately scaled DRAM using silicon migration technique by hydrogen anneling. In Proceedings of the 2017 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA, 2–6 December 2017; pp. 6.1–6.4. [Google Scholar]
  15. Schloesser, T.; Jakubowski, F.; van Kluge, J.; Graham, A.; Slesazeck, S.; Popp, M.; Baars, P.; Muemmler, K.; Moll, P.; Wilson, K. A 6F2 Buried Wordline DRAM Cell for 40nm and Beyond. In Proceedings of the 2008 IEEE International Electron Devices Meeting, San Francisco, CA, USA, 15–17 December 2008. [Google Scholar]
  16. Kang, J.; Cao, W.; Pal, A.; Pandey, S.; Kramer, S.; Hill, R.; Sandhu, G.; Banerjee, K. Computational study of gate-induced drain leakage in 2D-semiconductor field-effect transistors. In Proceedings of the 2017 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA, 2–6 December 2017; pp. 2.1–2.4. [Google Scholar]
  17. Hwang, Y.; Park, J.; Jin, G.; Chung, C. An Overview and Future Challenges of High Density DRAM for 20nm and Beyond. In Proceedings of the Extended Abstracts of the 2012 International Conference on Solid State Devices and Materials, Kyoto, Japan, 25–27 September 2012; The Japan Society of Applied Physics: Tokyo, Japan, 2012; pp. 586–587. [Google Scholar]
  18. Oh, C.W.; Yeo, K.H.; Kim, M.S.; Lee, C.-S.; Choi, D.U.; Kim, S.H.; Lee, S.-Y.; Kim, S.-M.; Choe, J.-D.; Lee, Y.K. Electrical Characterization of Partially Insulated MOSFETs with Buried Insulators under Source/Drain Regions. In Proceedings of the 30th European Solid-State Circuits Conference, Leuven, Belgium, 21–23 September 2004; pp. 233–236. [Google Scholar]
  19. Lee, M.J.; Cho, J.H.; Lee, S.D.; Ahn, J.H.; Kim, J.W.; Park, S.W.; Park, Y.J.; Min, H.S. Partial SOI Type Isolation for Improvement of DRAM Cell Transistor Characteristics. IEEE Electron Device Lett. 2005, 26, 332–334. [Google Scholar]
  20. Park, J.-M.; Sohn, S.-O.; Park, J.-S.; Han, S.-Y.; Lee, J.-B.; Kim, W.; Jeon, C.-H.; Kim, S.-D.; Kimt, Y.-P.; Lee, Y.-S. A Novel Body Effect Reduction Technique to Recessed Channel Transistor Featuring Partially Insulating Layer Under Source and Drain: Application to Sub-50nm DRAM Cell. In Proceedings of the 2007 IEEE International Electron Devices Meeting, Washington, DC, USA, 10–12 December 2007; pp. 903–906. [Google Scholar]
  21. Park, J.-S.; Park, J.-M.; Han, S.-Y.; Yamada, S.; Roh, Y.-H.; Park, D.-G. Bulk-FinFETs implementing insulating layer under source and drain for DRAM cell application. Electron Lett. 2008, 44, 824–825. [Google Scholar] [CrossRef]
  22. Ahn, S.J.; Jung, G.T.; Cho, C.H.; Shin, S.H.; Lee, J.Y.; Lee, J.G.; Jeong, H.S.; Kim, K. Novel DRAM cell transistor with asymmetric source and drain junction profiles improving data retention characteristics. In Proceedings of the 2002 Symposium on VLSI Technology, Honolulu, HI, USA, 11–15 June 2002; pp. 176–177. [Google Scholar]
  23. Park, H.; Choi, B. A study on the performance of metal-oxide-semiconductor-field-effect-transistors with asymmetric junction doping structure. Curr. Appl. Phys. 2012, 12, 1503–1509. [Google Scholar] [CrossRef]
  24. Cho, M.H.; Jeon, N.; Kim, T.Y.; Jeong, M.; Lee, S.; Hong, J.S.; Hong, H.S.; Yamada, S. An Innovative Indicator to Evaluate DRAM Cell Transistor Leakage Current Distribution. IEEE J. Electron Devices Soc. 2017, 6, 494–499. [Google Scholar] [CrossRef]
  25. Hurkx, G.A.M.; Klaassen, D.B.M.; Knuvers, M.P.G. A new recombination model for device simulation including tunneling. IEEE Trans. Electron Devices 1992, 39, 331–338. [Google Scholar] [CrossRef]
  26. Jin, S.; Lee, M.J.; Yi, J.H.; Choi, J.H.; Kang, D.G.; Chung, I.Y.; Park, Y.J.; Min, H.S. A new direct evaluation method to obtain the data retention time distribution of DRAM. IEEE Trans. Electron Devices 2006, 53, 2344–2350. [Google Scholar] [CrossRef]
  27. Kim, Y.K.; Lee, J.S.; Kim, G.; Park, T.; Kim, H.J.; Cho, Y.P.; Park, Y.J.; Lee, M.J. Partial Isolation Type Saddle-FinFET(Pi-FinFET) for Sub-30 nm DRAM Cell Transistors. Electronics 2019, 8, 8. [Google Scholar] [CrossRef] [Green Version]
Figure 1. 3D structure of a partial isolation type buried channel array transistor (Pi-BCAT).
Figure 1. 3D structure of a partial isolation type buried channel array transistor (Pi-BCAT).
Electronics 09 01908 g001
Figure 2. A cross-sectional view of y = 0 x–z in the doping distributions of five structures.
Figure 2. A cross-sectional view of y = 0 x–z in the doping distributions of five structures.
Electronics 09 01908 g002
Figure 3. Numerical comparison of on/off current in five structures.
Figure 3. Numerical comparison of on/off current in five structures.
Electronics 09 01908 g003
Figure 4. Linear (VDS = 0.1 V) and Log (VDS = 1.2 V) current graph of five structures.
Figure 4. Linear (VDS = 0.1 V) and Log (VDS = 1.2 V) current graph of five structures.
Electronics 09 01908 g004
Figure 5. Maximum transconductance (gm.max) and subthreshold slope (SS) comparison of five structures.
Figure 5. Maximum transconductance (gm.max) and subthreshold slope (SS) comparison of five structures.
Electronics 09 01908 g005
Figure 6. Vth and drain induced barrier lowering(DIBL) comparison of five structures.
Figure 6. Vth and drain induced barrier lowering(DIBL) comparison of five structures.
Electronics 09 01908 g006
Figure 7. Ion and Ioff according to the Lin in Pi-BCAT.
Figure 7. Ion and Ioff according to the Lin in Pi-BCAT.
Electronics 09 01908 g007
Figure 8. Electrostatic potential distribution according to the Lin of Pi-BCAT.
Figure 8. Electrostatic potential distribution according to the Lin of Pi-BCAT.
Electronics 09 01908 g008
Figure 9. Electric field-Z and band-to-band generation according to Lin = 6 nm and 12 nm of Pi-BCAT.
Figure 9. Electric field-Z and band-to-band generation according to Lin = 6 nm and 12 nm of Pi-BCAT.
Electronics 09 01908 g009
Figure 10. Changes in Ion and Ioff according to the silicon film of Pi-BCAT.
Figure 10. Changes in Ion and Ioff according to the silicon film of Pi-BCAT.
Electronics 09 01908 g010
Figure 11. Electrostatic potential distribution according to the silicon film thickness of Pi-BCAT.
Figure 11. Electrostatic potential distribution according to the silicon film thickness of Pi-BCAT.
Electronics 09 01908 g011
Figure 12. Change of Ion and Ioff according to Lin in Pi-BCAT at a gate depth = 70 nm.
Figure 12. Change of Ion and Ioff according to Lin in Pi-BCAT at a gate depth = 70 nm.
Electronics 09 01908 g012
Figure 13. Ion and Ioff change at a gate depth = 70 nm in Pi-BCAT according to the silicon film thickness.
Figure 13. Ion and Ioff change at a gate depth = 70 nm in Pi-BCAT according to the silicon film thickness.
Electronics 09 01908 g013
Figure 14. Change of Ion and Ioff according to Lin in Pi-BCAT at a gate depth = 90 nm.
Figure 14. Change of Ion and Ioff according to Lin in Pi-BCAT at a gate depth = 90 nm.
Electronics 09 01908 g014
Figure 15. Ion and Ioff change according to the silicon film thickness in Pi-BCAT at a gate depth = 90 nm.
Figure 15. Ion and Ioff change according to the silicon film thickness in Pi-BCAT at a gate depth = 90 nm.
Electronics 09 01908 g015
Figure 16. Optimal parameters (Lin and silicon film thickness) in Pi-BCAT at gate depths = 70 nm, 80 nm and 90 nm.
Figure 16. Optimal parameters (Lin and silicon film thickness) in Pi-BCAT at gate depths = 70 nm, 80 nm and 90 nm.
Electronics 09 01908 g016
Table 1. Electrical characteristics: Pi-BCAT with optimal parameters and asymmetry BCAT.
Table 1. Electrical characteristics: Pi-BCAT with optimal parameters and asymmetry BCAT.
Gate Depth Optimal Value70 nm80 nm90 nm
Lin5 nm5.5 nm5.75 nm
Silicon film thickness50 nm60 nm70 nm
Ion_Pi-BCAT2.21 × 10−6 A2.25 × 10−6 A2.31 × 10−6 A
Ioff_Pi-BCAT5.44 × 10−14 A1.98 × 10−14 A7.50 × 10−14 A
Ion_Pi-BCAT/Ion_Asymmetry BCAT80%80%80%
Ioff_Pi-BCAT/Ioff_Asymmetry BCAT38%33%38%
SSAsymmetry BCAT65.0 mV/dec65.4 mV/dec66.0 mV/dec
SSPi- BCAT66.225 mV/dec66.475 mV/dec66.725 mV/dec
gm.max:_Asymmetry BCAT7.22 × 10−6 A/V7.40 × 10−6 A/V7.64 × 10−6 A/V
gm.max:_Pi-BCAT5.74 × 10−6 A/V6.00 × 10−6 A/V6.16 × 10−6 A/V
DIBLAsymmetry_BCAT1.81 mV/V3.63 mV/V4.73 mV/V
DIBLPi-BCAT0.45 mV/V0.95 mV/V1.45 mV/V
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Lee, J.-s.; Park, J.-h.; Kim, G.; Choi, H.D.; Lee, M.J. Partial Isolation Type Buried Channel Array Transistor (Pi-BCAT) for a Sub-20 nm DRAM Cell Transistor. Electronics 2020, 9, 1908. https://doi.org/10.3390/electronics9111908

AMA Style

Lee J-s, Park J-h, Kim G, Choi HD, Lee MJ. Partial Isolation Type Buried Channel Array Transistor (Pi-BCAT) for a Sub-20 nm DRAM Cell Transistor. Electronics. 2020; 9(11):1908. https://doi.org/10.3390/electronics9111908

Chicago/Turabian Style

Lee, Jin-sung, Jin-hyo Park, Geon Kim, Hyun Duck Choi, and Myoung Jin Lee. 2020. "Partial Isolation Type Buried Channel Array Transistor (Pi-BCAT) for a Sub-20 nm DRAM Cell Transistor" Electronics 9, no. 11: 1908. https://doi.org/10.3390/electronics9111908

APA Style

Lee, J. -s., Park, J. -h., Kim, G., Choi, H. D., & Lee, M. J. (2020). Partial Isolation Type Buried Channel Array Transistor (Pi-BCAT) for a Sub-20 nm DRAM Cell Transistor. Electronics, 9(11), 1908. https://doi.org/10.3390/electronics9111908

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