Transcript 6.3 The Metal-Semiconductor FET

Chapter 6
Field Effect Transistors
6.1 Transistor Operation
6.2 The Junction FET
6.3 The Metal-Semiconductor FET
6.4 The Metal-Insulator-Semiconductor FET
6.5 The MOS Field-Effect Transistor
Field-effect Transistor(FET)
• junction FET (JFET)
• metal-semiconductor FET (MESFET)
• metal-insulator-semiconductor FET (MISFET)
metal-oxide-semiconductor FET (MOSFET)
FET v.s. BJT
Dominant features of p-n junction
1. The injection of minority carriers with forward bias → BJT
2. A variation of the depletion width W with reverse bias → FET
Comparison in terms of carrier type
• The FET is a majority carrier device, and is therefore often called a unipolar transistor.
• The BJT, on the other hand, operated by the injection and collection of minorpty carriers.
Comparison in terms of number of terminals
• Like its bipolar counterpart, the FET is a three-terminal device in which the current
through two terminal is controlled at the third.
Comparison in terms of control type
• Unlike the BJT, however, field-effect devices are controlled by a voltage at the third
terminal rather than by a current.
History of Transistors
1. FET was proposed first in 1930 by Lilienfeld.
2. Bardeen and Brattain invented the first bipolar transistor.
3. This major breakthrough was rapidly followed by Shockley’s extension of
the concept to the BJT.
4. The first MOSFET was demonstrated in 1960 by Kahng and Atalla.
6.1 Transistor Operation
6.2 The Junction FET
6.3 The Metal-Semiconductor FET
6.4 The Metal-Insulator-Semiconductor FET
6.5 The MOS Field-Effect Transistor
We can achieve large variations in vD by making small changes in vG. (voltage-controlled amplification)
We can switch from the bottom of the load line (iD = 0) to almost the top (iD ≈ E/R) by appropriate changes in vG.
(voltage-controlled switching)
6.1 Transistor Operation
6.2 The Junction FET
6.3 The Metal-Semiconductor FET
6.4 The Metal-Insulator-Semiconductor FET
6.5 The MOS Field-Effect Transistor
• The current ID flows through an n-type channel between two p+ regions.
• A reverse bias between these p+ regions and the channel causes the depletion regions to intrude into the n
material.
• The effective width of the channel can be restricted.
• The end of the channel from which electrons flow is called the source.
• The end toward which they flow is called the drain.
• The p+ regions are called gates.
• The voltage VG refers to the potential from each gate region G to the source S.
6.2.1 Pinch-off and Saturation
6.2.1 Pinch-off and Saturation
Without Gate Control
6.2.2 Gate Control
With Gate Control
• The effet of a negative gate bias VG is to increase the resistance of the channel and induce pinch-off at a lower
value of current.
• Pinch-off is reached at a lower value of VD (and therefore a lower ID) when a negative gate bias is applied. (see
next page)
W ( x  L)  [
2 ( VGD ) 1/ 2
]
qNd
(VGD negative)
(from Eq. 5-57)
W ( x  L)  a (pinch- off occurs)
2VP 1/2
[
] a
qNd
VP  VGD (pinch- off)  VG  VD
qa2 N d
VP 
2
(Vp at pinch-off: a positive number,
VG is zero or negative)
• Pinch-off is reached at a lower value of VD (and therefore a lower ID) when a negative gate bias is applied.
6.2.2 Current Voltage Characteristics
• Differential volume of the neutral channel
material: Z2h(x)dx
• Resistivity: 
l
dl
dx
R   ; dR    
A
A
Z 2 h( x )
Current I does not change with distance along the channel
dV  IdR
Z 2h( x) dVx
ID 

dx
The half-width of the channel at point x depends on VGx
h( x )  a  W ( x )  a  [
Z 2h( x) dVx
ID 

dx
qa2 N d
VP 
2
VGx  VG  Vx
2 (VGx ) 1/ 2
V V
]  a[1  ( x G )1/ 2 ]
qNd
VP
2Za

[1  (
Vx  0
Vx  VG 1/ 2
) ]dVx  I D dx
VP
VD 2 VG 3 / 2 2 VD  VG 3 / 2
I D  G0VP [  ( )  (
) ]
VP 3 VP
3
VP
VGx  0
VG  0
where Go 
Go is the conductance of the channel for negligible W(x), i.e., with no gate voltage and low values of ID
2aZ
L
6.2.2 Current Voltage Characteristics
Valid only up to pinch-off
I D  G0VP [
VD 2 VG 3 / 2 2 VD  VG 3 / 2
 ( )  (
) ]
VP 3 VP
3
VP
I D ( sat.)  G0VP [
 G0VP [
Go 
VG
2aZ
L
0
VD  VG  V p
VD 2 VG 3 / 2 2
 ( )  ]
VP 3 VP
3
VG 2 VG 3 / 2 1
 ( )  ]
VP 3 VP
3
V
VD
 1 G
VP
VP
The saturation current is greatest when VG is zero and becomes smaller as VG is made negative.
V
I D ( sat.)
 G0 [1  ( G )1/ 2 ] m utual transconductance
VG
VP
VG 2
I D ( sat.)  I DSS (1  ) , (VG negative)
VP
g m ( sat.) 
Experimentally
I DSS : saturateddrain current withVG  0
6.2.2 Current Voltage Characteristics
VG 2 VG 3 / 2 1
I D ( sat.)  G0VP [  ( )  ]
VP 3 VP
3
Go 
2aZ
L
Short channel effect
• Effect of : Electron velocity saturation at high fields may make this
assumption invalid (the assumption of Go constant is invalid)
• Effect of L: The effective channel length decreases as the drain voltage is
increased beyond pinch-off.
Therefore, the assumption of constant saturation current is not valid for vary
short-channel devices.
6.1 Transistor Operation
6.2 The Junction FET
6.3 The Metal-Semiconductor FET
6.4 The Metal-Insulator-Semiconductor FET
6.5 The MOS Field-Effect Transistor
6.3.1 The GaAs MESFET
(require doping)
By using GaAs instead of Si, a higher electron mobility is available. (Appendix III)
6.3.2 The High Electron Mobility Transistor (HEMT)
• The conductivity can be increased by increasing the doping in the channel and thus
the carrier concentration.
• However, increased doping also causes increased scattering by the ionized
impurities, which leads to a degradation of mobility.
Modulation doping
To grow a thin undoped well (e.g., GaAs) bounded by wider band gap, doped barriers
(e.g., AlGaAs).
The donors are in the AlGaAs rather than the GaAs, there is no impurity scattering of
electrons in the wwll.
Modulation doped field-effect transistor (MODFET)
High electron mobility transistor (HEMT) 2DEG
The donors in the AlGaAs layer are purposely separated from the interface by ~100 Å.
6.3.3 Short Channel Effects
vd 


1
vs
The saturated current
I D  qnvs A  qNd vs Zh
The saturated current follows the velocity saturation, and does not require a true
pinch-off in the sense of depletion regions meeting at some point in the channel.
The transconductance gm is essentially constant:
The ID-VD curves are more evenly spaced if constant saturation velocity dominates,
compared with the VG-dependent spacing for the long-channel constant-mobility
case.
6.3.3 Short Channel Effects
1. The saturated current follows the velocity saturation, and does not require a true
pinch-off in the sense of depletion regions meeting at some point in the channel.
• Most devices operate with characteristics intermediate between the constant
mobility and the constant velocity regimes. Approximation required.
2. Another important short-channel effect is the reduction in effective channel
length pinch-off as the drain voltage is increased.
• This effect is not significant in long-channel devices, since the change in L
due to intrusion of the depletion region is a minor fraction of the total
channel length.
• In the short-channel devices, however, the effective channel length can be
substantially shortened, leading to a slope in the saturated I-V characteristic
that is analogous to the Early (base-width narrowing) effect in BJTs.
6.1 Transistor Operation
6.2 The Junction FET
6.3 The Metal-Semiconductor FET
6.4 The Metal-Insulator-Semiconductor FET
6.5 The MOS Field-Effect Transistor