• 沒有找到結果。

Voltage scaling and temperature effects on drain leakage current degradation in a hot carrier stressed n-MOSFET

N/A
N/A
Protected

Academic year: 2021

Share "Voltage scaling and temperature effects on drain leakage current degradation in a hot carrier stressed n-MOSFET"

Copied!
5
0
0

加載中.... (立即查看全文)

全文

(1)

Voltage Scaling and Temperature Effects on Drain Leakage Current

Degradation in a Hot Carrier Stressed n-MOSFET

Tahui Wang, C.F. Hsu, L.P. Chiang. N.K. Zous. T.S. Chao* and C.Y. Chang Department of Electronics Engineering, National Chiao-Tung University

*National Nan0 Devices Lab.,

fax:886-3-5724361 ;email: wang@ jekyll.ee.nctu.edu.tw Hsin-Chu, Taiwan, R.O.C.

ASSTRACT

Drain leakage current degradation at zero V in a hot carrier stressed n-MOSFELT is measured and modeled. h e dependences of drain leakage current on supply voltage and temperature are characterized. In modeling, various drain leakage current mechanisms including drain-to-source subthreshold leakage current, band-to-band tunneling current and interface trap assisted leakage ciment are taken into account. Our result shows that interface trap induced leakage current appears to be a dominant drain 1,:akage mechanism as the supply voltage is scaled below 3.0V. Drain leakage current degradation by orders of magnitude has been observed due to hot carrier stress.

blTRODUCnON

The reduction of drain leakage current at zero Vgs has been a major concem in CMOS device scaling. Gate induced drain leakage (GIDL) current resulting from band-to-band tunneling has been recognized as one of the major drain leakage mechanisms in thin-oxide MOSFETs [l]. Recently, hot carrier stress effects on device degradation have received much interest [2-61. However, most of the studies concentrate on stress induced on-state drain current reduction [2,3]. The stress effect on off-state drain leakage current degradation has not received as much attention [4,7]. The stress effect on drain leakage current is attribuud to the creation of oxide trapped charge and interface traps. The build-up of negative oxide charge shifts the device flat-band voltage and results in an enhancement of band- to-band tunneling current. In addition, the generated interface traps can introduce an additional trap-assisted leakage mechanism [8]. At scaled supply voltages, while band-to-band tunneling can be greatly alleviated, the trap-assisted current may appear to be a dominant leakage mechanism. Furthermore, experimental results showed that the trap-induced leakage current exhibits a dependence on temperature [9]. In a certain bias range, the lcakage current becomes much aggravated as temperature rises and thus may have impact on a DRAM refresh time. In this woik, it is our intention to investigate the hot carrier stress effect on drain leakage current degradation in thin- oxide MOSFETs. The drain leakage current mechanisms at different supply vlaltages and temperatures are characterized and modeled.

DRAIN

LEAKAGE CURRENT MECHANISMS

Various drain leakage paths in a stressed MOSFET are considered in our model. The trap-independent leakage mechanisms include band-to-band tunneling current

(Im)

and drain-to-source subthreshold current (IS),

IBB = M?exp(-B/Et) ( 1 )

Is

=

roexp(-ev,,)

(2)

nkT

where the parameters A and B are d e f i e d in Ref. [lo]. Et denotes the total Si surface field. Drain junction leakage current is small in this work and can be neglected. Fig. 1 illustrates these two components in the lateral direction and in the vertical direction respectively.

+

Is

drain

L

(a)lateral direction

gate^^

substrate (b)vertical direction

Fig. 1 Illustration of subthreshold leakage current

(Is)

and band- to-band tunneling current

(lm)

(a) in the lateral direction and (b) in the vertical direction.

The interface trap assisted leakage mechanisms are the Shockley-Read-Hall current ( ~ J s R H ) . thermionic-field emission current

(a&.

and sequential tunneling current (&TAT). A complete trap-assisted drain-to-substrate leakage path at the Si/SiO2 surface is formed by hole emission from traps to the valence band and electron emission from traps to the conduction band. Both electron emission and hole emission are carried out via either thermionic emission or field emission. The carrier transition processes are drawn in Fig. 2. In the figure, Ge and Gh

stand for electron and hole thermionic emission rates and Te and

Th denote electron and hole tunneling rates. Each carrier transition rate is formulated as follows,

(2)

(9

where on and

up

are electron and hole capture cross sections, &i and stand for the intrinsic Fermi-level and trap energy. fv ,

fr

and f , are the electron occupation factors in the valence band, trap states and in the conduction band, respectively. ns and ps are electron and hole concentrations at the Si surface, which are calculated from a two-dimensional device simulation.

re

and 2 h

are electron and hole tunneling times from the WKB approximation. In a steady-state, the trap occupation factor

ft

can be evaluated from the equality

Ge+Te=Gh+Th

with fc=O and fv= 1. The three trap-assisted leakage current components are therefore expressed in the following [8],

where dL is the width of the interface trap ( W i t ) region and

W

is the channel width. The total trap-assisted leakage current d l d is the summation of the above three components,

(a) lateral direction

(b)

vertical direction

lateral direction and (b) in the vertical direction. Note that

TA

only occurs in the lateral direction.

The temperature-dependent parameters used in the calculation are bandgap (&g), thermal velocity (vth) and intrinsic concentration (ni ). They are given below [ 113,

The fielddependent parameters are Ze and Zh,

where

roc

and rOv are effective transit times in the conduction band and in the valence band, is the surface field in the lateral direction and other variables have their usual definitions. Note that the hole tunneling

Th

occurs only in the lateral direction. Thus,

Th

is a function of only a lateral field whereas the tunneling

process 1~ and

Te

are dependent on a total field.

DEVICE

CHARACTEREATTON

The test device is a 0.35p.m n- T with source/drain extension. The gate oxide thicknes ut 40A and the gate width is 100pm. The device was subject to maximum substrate current stress Vg8=2V and Vh=4.5V for 3000 seconds. The pre- stress and post-stress Id -Vgs characteristics are shown in Fig. 3 (T=292K) and in Fig. 4 (T=353K), respectively. Under the stress condition, interface trap generation is almost saturated and oxide charge creation is minimal [7]. Interface trap generation is

evidenced by the change of the subthreshold swing in Figs. 3 and 4. The

DIBL

effect is not significant in the device.

-1.0 -0.5 0.0 0.5 1.0 1.5

Gate-to-source

Bias,

Vgs

(V)

(3)

10.' n

E

2-

10.'

E"

z

n

w

c)

8

10-l1 10." .I ed tcr -1.0 -0.5 0.0 0.5 1.0 1.5

Gate-to-source Bias, Vgs

(V)

Fig. 4 Measured Ire-stress and post-stress Id -Vss characteristics at V,=1.5\' and 3.0V, T=353K.

RESULTS

AND

DISCUSSIONS

The dependence of pre-stress and post-stress drain leakage currents on supply voltage (VdJ is measured in Fig. 5. The leakage current enhancement factor, defined as the ratio of the post-stress drain leakage to the pre-stress drain leakage at T=292K, is shown in Fig. 6. The trap effect on the leakage current enhancement becomes particularly pronounced around a supply voltage of 2.2V. An enlargement of the drain leakage current by a factor of 18 is obstmed at T=292K. Various drain leakage current components lM, 158 and the trap induced d l d are plotted in Fig. 7. The solid lines are the measured result and the circles represent the calculated result.

-

pre-stress

-pI

/' 0 1 1.0 1.5 2 0 2.5 3.0 3.5

Supply Voltage (V)

Fig. 5 Measured ckain leakage current characteristics before and after stresf. at two different temperatures, T=292K and T=353K.

120

T=353K

1.3 1.7 2.1 2.5 2.9 3.3

Supply Voltage (V)

Fig. 6 The ratio of the post-stress drain leakage current to the pre-stress (T=292K) drain leakage current versus supply voltage.

The interface trap density

Nit

used in the calculation is 1 . 4 ~ 1 0 ~ ~ c m - ~ and the length of the trap distribution (AL) is 400A.

a,,

and 0, in Eqs. (3) and (4) are 1015cm2[11]. The following features are observed. At a large supply voltage (Vd3.OV). the band-to-band tunneling current (I@) manifests itself as a dominant mechanism even in a stressed device. In addition. the trap induced component d l d exhibits a weaker field dependence than the 1~ due to a smaller tunneling barrier from traps to the conduction band (electron tunneling) or to the valence band (hole tunneling). As a result, as the supply voltage scales, the

Im

drops more quickly and the A I d appears to be a major drain leakage mechanism in a stressed device. When the V u is further reduced, the tunneling effect becomes unimportant. The enhancement of the drain leakage is achieved mainly through the

SRH

component. Therefore, the enhancement factor in Fig. 6 peaks around 2.2V. At T=353K, the enhancement factor in Fig. 6

is significantly increased up to 110. The temperature effect is apparent at a lower V u The enhancement factor at V&=1.3V increases from about 5 at T=292K to 80 at T=353K. The [M, fs

and the A I d at T=353K are shown in Fig. 8.

10"

10-13

1.3 1.7 2.1 2 5 2.9 3.3

Supply Voltage (V)

Fig. 7 Various drain leakage current components at T=292K from measurement (solid lines) and calculation (full circles).

(4)

10-8

T=353K

-

measurement 10-8

T=353K

E .111

E

1 6 "

n

Is 1.3 1.7 2.1 2.5 2.9 3.3

Supply Voltage (V)

Fig. 8 Various drain leakage current components at T=353K from measurement (solid lines) and calculation (full circles).

d l d reprensents the trap-induced drain leakage.

Furthermore, the components in the a d are analyzed in Fig. 9 (T=292K) and in Fig. 10 (T=353K). In Fig. 9, the trap- induced drain leakage is dictated by the &AT for vdd >1.7v, by the

d l ~ ~

for 1.7v?V&.5v, and by the NSRH for 1.SVrVdd In Fig. 10, the thermally related components AZTF and NSRH are more prominent. As the supply voltage scales, the field dependent cunent component (A~TAT) can be greatly alleviated, while the temperature dependent components ( ~ T F and AZSRH)

become dominant drain leakage mechanisms.

1.3 1.7 2.1 2.5 2.9 3.3

Supply Voltage (V)

Fig. 9 Calculation of various interface trap induced drain leakage current components at T=292K.

The vertical field and the lateral field effects on drain leakage current are also assessed. Our calculation reveals that the band-to- band tunneling, current is mainly contributed by the vertical field whereas the trap-assisted tunneling is mostly determined by the lateral field due to the dependence of T h (hole tunneling) only on the lateral field.

1.3 1.7 2.1 2.5 2.9 3.3

Supply Voltage (V)

Fig. 10 Calculation of various interface trap induced drain leakage current components at T=353K.

CONCLUSIONS

The band-to-band tunneling current is found to be a dominant drain leakage mechanism at V e 3 . 3 V in the current device structure. At a reduced supply voltage, the trap-induced current appears to be the major leakage component. We have observed that hot carrier stress can degrade the drain leakage current by orders of magnitude. While the vertical field has a larger effect on the band-to-band tunneling current, the trap- induced leakage is mostly caused by lateral field enhanced tunneling. The reduction of lateral field is necessary to suppress the trap-induced drain leakage in deep submicron devices.

ACKNOWLEDGEMENT

Financial support from National Science Council,

ROC,

is gratefully acknowledged. The authors also would like to express thanks to Dr. Chimoon Hung at Macronix Inc. for many fruitful discussions.

REFERENCES

J. Chen. T.Y. Chan, P.K. KO, and C. Hu, 'Subbreakdown Drain Leakage Current in MOSFET,' IEEE Electron Device Lett. Vol. EDL-8, pp. 515-517. 1987.

T. Tsuchiya, "Trapped-electron and Generated Interface Traps in Hot Electron Induced MOSFET Degradation," IEEE Trans. Elect. Dev.

,

Vol. ED-34, pp. 2291-2296,

1987.

T. Wang, C. Huang, P. C. Chou,

S. S.

Chung and T. E. Chang, "Effects of Hot Carrier Induced Interface State Generation in Submicron LDD MOSFET's. " IEEE Trans.

Efect. Dev. , Vol. ED-41, pp. 1618-1622, 1994. H. Sasaki, M. Saitoh, and K. Hashimoto, "Hot-carrier Induced Drain Leakage Current in n-channel MQSFET,"in IEDM Tech. Dig., pp. 726-729, 1987.

C. Duvvury, D. I. Redwine, and H. J. Stiegler, "Leakage Current Degradation in N-MOSFETs due to Hot-electron Stress,'' IEEE Electron Device Lett., Vol. EDL-9, pp.579-58 1, 1988.

G. Q.

Lo,

A. B. Joshi, and D-L, Kwong, "Hot-carrier- stress Effects on Gate-induced Drain Leakage Current in nchannel MOSFET's," IEEE Electron Device Lett., Vol. A. Frommer, M.

R.

Pinto, and J. D. Bude, "Two-Stage Leakage Degradation in Sub-Micron MOSFET Technology", Symp. on VLSI Tech., pp.164-165, 1996

(5)

[8] T. Wang, T. E. Chang, and Chimoon Hung, "Interface Trap Induced Thermionic and Field Emission Current in Off-State MOSI'ETs" in IEDM Tech. Dig., pp.161-164,

1994.

[9] C. T. Wang, Hot Carrier Design Consideration for MOS Devices and Circuits, Van Nostran Reinheld, 1992 [lo] T. Y. Chan, J. (:hen, P. K.

KO,

and C. Hu. 'The Impact

of Gate-induced Drain Leakage Current on MOSFET Scaling," in IEllM Tech. Dig., pp. 721-724, 1987. [ll]

S. M.

Sze. Sc'micondvctor Device Physics and

數據

Fig.  1  Illustration of  subthreshold leakage current  (Is)  and band-  to-band tunneling current  (lm)  (a) in the lateral direction  and  (b) in the vertical direction
Fig.  3 Measured pre-stress and post-stress  Id -Vgs characteristics
Fig.  6  The ratio  of  the  post-stress  drain  leakage current  to  the  pre-stress  (T=292K) drain  leakage current versus  supply  voltage
Fig.  10  Calculation of various interface trap induced drain  leakage current components at T=353K

參考文獻

相關文件

While Scheme-KGs may have flexibility in maintaining the current arrangements in learning and teaching activities 1 , they are required to appoint teachers to meet the

Put the current record with the “smaller” key field value in OutputFile if (that current record is the last record in its corresponding input file) :. Declare that input file to be

To reduce the leakage current related higher power consumption in highly integrated circuit and overcome the physical thickness limitation of silicon dioxide, the conventional SiO

Take a time step on current grid to update cell averages of volume fractions at next time step (b) Interface reconstruction.. Find new interface location based on volume

9 The pre-S1 HKAT is conducted in all secondary schools in July every year to assess the performance of students newly admitted to S1 in Chinese Language, English Language

Workshop of Recent developments in QCD and Quantum field theories, 2017

In section29-8,we saw that if we put a closed conducting loop in a B and then send current through the loop, forces due to the magnetic field create a torque to turn the loopÆ

The thesis uses text analysis to elaborately record calculus related contents that are included in textbooks used in universities and to analyze current high school