ECE442 Spring2007 3.

PNJunctionsandDiodes

Jose E. Schutt-Aine Electrical & Computer Engineering University of Illinois [email protected]

PN Junction

When a p material is connected to an n-type material, a junction is formed Holes from p-type diffuse to n-type region Electrons from n-type diffuse to p-type region Through these diffusion processes, recombination takes place Some holes disappear from p-type Some electrons disappear from n-type

A depletion region consisting of bound charges is thus formed Charges on both sides cause electric field potential = Vo
JoseE.SchuttAine ECE442 Spring2007

PN Junction
Potential acts as barrier that must be overcome for holes to diffuse into the n-region and electrons to diffuse into the p-region Open circuit: No external current

Junction built-in voltage

N N Vo = VT ln A 2 D ni
For Si, Vo is typically 0.6V to 0.8V Charge equality in depletion region gives:
qx p AN A = qxn AN D

A: cross-section of junction xp: width in p side xn : width in n side

PN Junction
xn N A = xp ND 2 s 1 1 = xn + x p = + Vo q N A ND

Wdep

s : silicon permittivity s = 11.7 o = 1.04 108 F/m

JoseE.SchuttAine ECE442 Spring2007

PN Junction under Reverse Bias

When a reverse bias is applied Increase of space charge region Diffusion current decreases Drift current remains constant Barrier potential is increased A steady state is reached When Increase in depletion-layer voltage above Vo will appear as an external voltage between the diode terminals

Under reverse bias the current in the diode is negligible

Depletion Layer Stored Charge

qj = q

N AND AWdep N A + ND

The total voltage across the depletion layer is Vo + VR

Wdep =

2 s 1 1 + (Vo + VR ) q N A ND

JoseE.SchuttAine ECE442 Spring2007

Depletion Capacitance
Cj = dq j dVR
VR =VQ

Q is bias point VR is reverse voltage

Cj =

s A
Wdep

C jo 1+ VR Vo
Cj = C jo VR 1 + Vo
m

sq N A ND 1 C jo = A 2 N A + N D Vo

m is the grading coefficient and depends on how the concentration varies from the p side to the n side 1/3 <m <1/2
For an abrupt junction, m=0.5
JoseE.SchuttAine ECE442 Spring2007

PN Junction
Summary
Reverse bias voltage is applied to pn junction Transient occurs during which depletion capacitance is charged to new bias voltage reverse After transient: steady-state reverse current = IS-ID (ID is very small) current ~ IS ~10-15 A

JoseE.SchuttAine ECE442 Spring2007

Diode I-V Relationship

Forward

0.5V

0.7V

Reverse
Breakdown

Breakdown
Electric field strong enough in depletion layer to break covalent bonds and generate electron-hole pairs. Electrons are then swept by E-field into the nside. Large number of carriers for a small increase in junction voltage

JoseE.SchuttAine ECE442 Spring2007

Forward-Biased Junction Carrier Distribution

pn(xn)
Depletion Region

N Region

np(-xp)

Excess concentration

pn(x) pno
Thermal Equilibrium Value

np(x)
npo

-xp

xn

NA >> ND Barrier voltage is now lower than Vo In steady state, concentration profile of excess minority carriers remains constant

JoseE.SchuttAine ECE442 Spring2007

10

Forward-Biased PN Junction
Distribution gives rise to diffusion current ID above the value of IS

pn ( xn ) = pno eV / VT
xn is at junction
Excess minority carriers

pn ( x) = pno + pn ( xn ) pno e

( x xn ) / L p

Lp: diffusion length of holes in n-type Si L p = D p p

Dp: diffusion constant for holes p: excess minority carrier lifetime

1 ns < p < 10,000 ns

JoseE.SchuttAine ECE442 Spring2007

11

Forward-Biased PN Junction
Current density due to hole injection

Lp Current density due to free electron

Jp = q

Dp

pno eV / VT 1

Ln: diffusion length of electrons in p region

)
n po = ni2 / N A

since

pno = ni2 / N D

and

Dp Dn V / VT I = Aqni2 + 1 e L N p D Ln N A

)
12

JoseE.SchuttAine ECE442 Spring2007

Forward-Biased PN Junction
Diode equation:
I D = I S eV / VT 1

Dp Dn I S = Aqni2 + L p N D Ln N A

2 since ni is a strong function of temperature; thus Is is a strong function of temperature

JoseE.SchuttAine ECE442 Spring2007

13

Diode Circuits
Vout = VD I D = I S eVD / VT 1

VS = RI D + VD = RI D (VD ) + VD
Nonlinear transcendental system
ID Vs/R

Use graphical method

Diode characteristics Load line (external characteristics)

Vout

VS

VD

Solution is found at itersection of load line characteristics and diode characteristics

14

Diode Circuits Iterative Methods

Vout = VD

x ( k +1) = x ( k ) f '( x ( k ) )

f ( x(k ) )

f (VD ) =

VD VS + I S eVD / VT 1 = 0 R 1 I f '(VD ) = + S eVD / VT R VT

VD( k ) VS V
( k +1) D

=V

(k ) D

+ IS e

V ( k ) / VT
D

( 1 I S VDk ) / VT + e R VT

( Where VDk ) is the value of VD at the kth iteration

Procedure is repeated until convergence to final (true) value of VD which is the solution. Rate of convergence is quadratic.
JoseE.SchuttAine ECE442 Spring2007

15

Diode Circuits - Rectification

Rectification with ripple reduction.

C must be large enough so that RC time constant is much larger than period

JoseE.SchuttAine ECE442 Spring2007

16

Example
Find the barrier voltage across the depletion region of a silicon diode at T = 300 K with ND=1015/cm3 and NA=1018/cm3. Use
N AND Vo = VT ln 2 ni

@ 300K, ni = 1.5 1010 /cm3 VT = 0.026 V

1018 1015 1013 Vo = o = 0.026ln = 0.026ln (1.5 )2 1020 2.25

JoseE.SchuttAine ECE442 Spring2007

17

Example
Two diodes are connected in series as shown in the figure with Is1 =10-16 A and Is2 =10-14 A. If the applied voltage is 1 V, calculate the currents ID1 and ID2 and the voltage across each diode VD1 and VD2.

The diode equations can be written as: VD 1 VD 2 I S1 I I D1 = I S 1eVD1 / VT I D 2 = I S 2eVD 2 / VT e VT = D1 = 1 IS 2 I D2 I from which VD1 VD 2 = VT ln S 1 = 0.12 IS 2 Using KVL, we get VD1 + VD 2 = 1 from which VD 2 = 0.44 V and VD1 = 0.56 V

I D1 = 1016 e0.56 / 0.026 = 0.22 A=I D 2

18