3.4 Charge Pumping Applicability on Ge MOSFET
3.4.4 Limitations of Charge Pumping on Ge
exhibits lower Dit value. Symmetric distribution is attained and consistent with conductance method again. Unfortunately, we are not able to obtain the midgap profile. In order to make the approach the midgap, tr/tf much longer than 100ns (at least 1μs) should be applied so that low frequency measurement (lower than 50 kHz) is needed. For low frequency measurement practical, annealing condition must be optimized and our 100μm2 junction area should be smaller.
3.4.4 Limitations of Charge Pumping on Ge
The biggest obstacle using charge pumping on Ge is that junction leakage and noise can jeopardize charge pumping measurements, especially for large and defective Ge junctions.
The small voltage difference between S/D and bulk can be sufficient to cause leakage currents to dominate charge pumping currents. For charge pumping measurements to be effective, it is preferable to have well optimized and small junctions.
There exist issues for charge pumping at low temperatures related to electrical contacts.
The series resistance might increase at low temperature due to freeze-out of Schottky contacts, while Charge pumping measurements are sensitive to excessive series resistance.
50
3.5 Conclusions
In Chapter 3, we investigated the effect of FGA on Ge p+n junction and device characteristics. On/off ratio of our p+n junction and p-FET reached 4.8 orders and 3 orders respectively (500°C 30s dopant activation, W/L = 100μm/10μm), with better on/off ratio (3.3 orders) and subthreshold swing (170mv/dec) obtained after FGA. Charge pumping measurement was done to obtain the midgap Dit information, and after FGA between
and is 4.2×1011cm-2eV-1 for 500°C GeO2 passivation and 1.1×1012cm-2eV-1 for no passivation sample. Besides, were 5.4×10-16 cm2 and 1.1×10-15 cm2 for samples with and without GeO2 passivation respectively. Both and were consistent with the results derived from conductance method applied to MOSCSPs in Chapter two, reconfirming the validity of conductance method.
Pros and cons of adding the GeO2 layer were summarized according to our experimental data. Lower Dit value verified from either charge pumping or gated diode measurement made the mobility higher for GeO2 passivation sample, while it suffered from more severe carrier-trapping due to border traps at the GeO2/Al2O3 interface and more severe subthreshold swing degradation.
51
References (Chapter 3)
[1] G. Wang, R. Loo, E. Simoen, L. Souriau, M. Caymax, M. M. Heyns, and B. Blanpain,
"A model of threading dislocation density in strain-relaxed Ge and GaAs epitaxial films on Si(100)", Appl. Phys. Lett., vol. 94, p. 102115, 2009.
[2] R. Xie, T. H. Phung, W. He, Z. Sun, M. Yu, Z. Cheng and C. Zhu, ―High Mobility High-k/Ge pMOSFETs with 1 nm EOT -New Concept on Interface Engineering and Interface Characterization‖, IEDM Tech. Dig, pp. 393-396, 2008.
[3] S. Zafar, Y. H. Kim, V. Narayanan, C. Cabral Jr., V. Paruchuri, B. Doris, J. Stathis, A.
Callegari and M. Chudzik, ―A Comparative Study of NBTI and PBTI (Charge Trapping) in SiO2/HfO2 Stacks with FUSI, TiN, Re Gates,‖ Tech. Dig. VLSI Symp., 2006.
[4] G. Groeseneken, H. E. MAES, N. BELTRAN, AND R. F. KEERSMAECKER, ―A Reliable Approach to Charge-Pumping measurements in MOS Transistors,‖ IEEE Trans.
Electron Devices, vol. 31, p. 42, 1984.
[5] K. Martens, B. Kaczer, T. Grasser, B. De Jaeger, M. Meuris, H. E. Maes, and G.
Groeseneken, ―Applicability of Charge Pumping on Germanium MOSFETs,‖ IEEE
Electron Devices Lett., 2006.
[6] R. Xie, N. Wu, C. Shen, and C. Zhu, "Energy distribution of interface traps in germanium metal-oxide semiconductor field effect transistors with HfO2 gate dielectric and its impact on mobility", Appl. Phys. Lett., vol. 93, p. 083510, 2008.
52
Fig. 3.1 Process flow of Ge p-MOSFETs and their device structure.
53
(a)
(b)
Fig. 3.2 I–V characteristics of p+n junctions activated at 500°C of the two samples, before and after performing FGA. (a) No passivation (b) GeO2 passivation
-2 -1 0 1 2
10
-510
-310
-110
110
3Current density, J(A/cm2 )
Voltage (volt)
no passivation
500oC 30s dopant activation before FGA, n=1.22 after FGA, n=1.22
-2 -1 0 1 2
10
-510
-310
-110
110
3Current density, J(A/cm2 )
Voltage (volt)
GeO2 passivation before FGA, n=1.24 after FGA, n=1.32
54
(a)
(b)
Fig. 3.3 Series resistance extracted from p+n junctions of the two samples, before and after performing FGA. (a) No passivation (b) GeO2 passivation
0.000 0.010 0.020 0.030 0.0
0.2 0.4 0.6 0.8
no passivation with FGA, rs=34.6
w/o FGA, rs=16.9
I=I0,qnre^(q(V-Irs)/nkT) I/gd =nkT/q+Irs
slope=rs
I/g d (V)
Current (A)
55
0.00 0.05 0.10 0.15 0.20 0.25 0.30
0
56
Fig. 3.5 Effects of FGA at 300°C on the series resistance from Terada and Muta method. (a) No passivation. (b) GeO2 passivation.
57 passivation. (b) GeO2 passivation.
58
1 0 -1 -2 -3 -4 -5
0.01 0.1 1 10 100
SS=169.9mV/dec SS=271.5mV/dec
with FGA, VD=-0.1V, -2.1V
no passivation GeO2 passivation W/L=100m/10m
Drain Current, I D (A/m)
Gate Voltage, V
G(V)
Fig. 3.7 Comparison of ID-VG characteristics between no passivation and GeO2 passivation sample.
59
0.0 -0.5 -1.0 -1.5
-0.84 -0.88
-0.92 no passivation
with FGA V
D=V
S=-0.2V I
gen,s=6.6E-8A
Drain/Source Current (A)
Gate Voltage, V
G(V)
(a)0.0 -0.5 -1.0 -1.5
-0.96 -0.98 -1.00
-1.02 GeO
2passivation
with FGA V
D=V
S=-0.2V I
gen,s=4.0E-8A
Drain/Source Current (A)
Gate Voltage, V
G(V)
(b)Fig. 3.8 Gated-diode measurement detects the interface state density roughly. (a) No passivation. (b) GeO2 passivation.
60
Fig. 3.9 Split-CV before FGA is measured. (a) No passivation. (b) GeO2 passivation.
61
0.0 0.2 0.4 0.6 0.8
0 100 200 300 400
1.3X 1.7X
Si universal curve
2008 IEDM GeO2 passivation no passivation
Mobility(cm2 /Vs)
Electric Field (MV/cm)
3Xw/o FGA
Fig. 3.10 Effective mobility VS effective electric field is plotted before FGA with the highest published hole mobility up to date.
62
Fig. 3.11 Effects of FGA at 300°C on the effective mobility of GeO2 passivation Ge p-MOSFETs. (a) Cgc-VG plot. (b) Qinv-VG plot. (c) gd-VG plot. (d) μeff - Qinv plot.
63
Fig. 3.12 Reliability is compared between no passivation and GeO2 passivation sample under static stress after FGA. (a) Charge trapping behavior. (b) Ratio of early traps. (c) Subthreshold swing degradation.
64
Fig. 3.13 Fixed amplitude charge pumping method was applied on the Ge p-MOSFFET after FGA. (a) No passivation. (b) GeO2 passivation.
65
1E-7 1E-6 1E-5 1E-4
-0.03 -0.02 -0.01 0.00
f =500KHz no passivation
specific t
r/t
f=1.29E-5s
Icp(A/cm2 )Transition Time (s)
(a)1E-7 1E-6 1E-5 1E-4
-0.012 -0.009 -0.006 -0.003 0.000
f =500KHz GeO
2passivation specific t
r/t
f=2.63E-5s
Icp(A/cm2 )Transition Time (s)
(b)Fig. 3.14 Icp versus transition time are plotted after FGA, with specific tr/tf extracted. (a) No passivation. (b) GeO2 passivation.
66
Fig. 3.15 Positions of electron and hole emission level of GeO2 passivation sample as a function of tr/tf at different temperatures are calculated.
10
-810
-710
-610
-50.0
0.1 0.2 0.3 0.4 0.5 0.6 0.7
0.43eV
80k 200k
Eem,h
E- Ev (eV)
transition times (s) Eem,e
300k
0.28eV
=5.4E-16cm
267
D
itdistribution in Ge band gap
Dit(cm-2 eV-1 )
Et-Ei (eV)
Fig. 3.17 Dit distribution in the Ge band gap of the two samples is depicted.
2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5
68
Chapter 4
Inversion-Mode Ge n-MOSFET with Atomic-Layer-Deposited Al 2 O 3 Gate Dielectrics
4.1 Introduction
Ge has become of great interest as a channel material for future technology nodes, owing to its high bulk electron and hole mobility. Although high-k/Ge p-MOSFET with mobility 3X higher than Si universal curve has been reported recently [1], the performance of promising Ge n-MOSFETs is still unsatisfactory with low extracted electron mobility. The mechanisms of n-MOSFET degradation can be explained in terms of fast trapping at Ge/GeO2 interface, slow trapping by GeO2/Al2O3 border traps and parasitic S/D series resistance [2]. Passivation of Ge interface and activation of n-type dopants have been considered the major challenges to achieve high performance Ge CMOS.
In this chapter, SiO2/GeO2 isolation is demonstrated to be better than the SiO2 isolation, with the underlying leakage paths and mechanisms analyzed. Then, the device electrical characteristics of Ge n-MOSFET with different GeO2 thickness are discussed, including series resistance, subthreshold swing, and mobility. The interface qualities for different samples are characterized by charge pumping and gated-diode measurement, in order to find the correlation between interface quality and device characteristic. Reliability issues for GeO2 are investigated through applying constant voltage stress, and it is confirmed that low conduction band offset of GeO2 on Ge is a potential problem.
69