Jpl Iitb.pres2

8/10/2019 Jpl Iitb.pres2

1/28

Gate Carrier Injection and NC-Non-

Volatile Memories

Jean-Pierre Leburton

Department of Electrical and Computer

Engineering and Beckman Institute

University of Illinois at Urbana-Champaign

Urbana, IL 61801, USA


2/28

Hot Carrier Effects in MOSFETs

High-field/non-linear transport

vx

kBT

c

Fx

Long tail energy

distribution

E =(1 / 2)m*v2

*

*After R.S. Muller and T.I Kamins, DEIC, Wiley, 3d ed.

f(v)

J.P. Leburton, IWSG-2009, IITB, India


3/28

Hot Carrier Effects: Substrate Current*

Ec

EG

Ev

SS

SS

S

S

SS

SE>EG

e

ee

h+

Impact Ionization

I-V characteristics

*After R.S. Muller and T.I Kamins, DEIC, Wiley, 3d ed.J.P. Leburton, IWSG-2009, IITB, India


4/28


5/28

Tunneling Injection into the Gate*

Direct Tunneling

Fowler-Nordheim TunnelingTrap-Assisted Tunneling

Electron trapping in SiO2 Hole trapping in SiO2

*After Y.Taur and T.H. Ning, FMVD, Cambridge, 2d ed.

JFN !exp(!4 2qm *

3"Fox"ox

3/ 2 )

Fox

!ox

VT- Degradation

Dissipation (gate leakage)

JTun

= JTun

(tox

,!ox)


6/28

Injection into Floating Gates

n-channel

p-channel

Drain-avalanche

(impact ionization)

Injection by channel hot electrons (CHE)

No CHEbecause largeroxide barrier


After R.S. Muller and T.I Kamins, DEIC, Wiley, 3d ed.


7/28


8/28

Flash Memory Device: Basic Operation*

ETOX: Hot electron-tunneling combined VT-shift

But leakage throughdefects!!!

* A. Thean and J.P.Leburton, IEEE Potentials, 21(4) 35, (2002)

!FG electrically disconnected

!Data stored in form of charge packages

!Transport mechanisms (CHE)!FN tunnelling (oxide damage)

!Memory cells altered individually

!Data storage sensed by conductance

!Non-volatile storage

!Down scaling X retention time



9/28


10/28

Nanocrystal Memories*

* S. Tiwari et al. IEDM Tech Dig., 521, Dec. 1995.

**

** Courtesy Motorola Inc.

Single electron charging***

"VG=e/C; C:NC capacitance

NC memory operation principle*

NC memory device

structure

"E


11/28

NC Memory Device: QM Modeling*

Schroedinger Equation (effective mass approx.)

Simulated structure Crystallographic orientations

!!2

2("x ,"y ,"z )

mxx!1("r ) mxy

!1("r ) mxz

!1("r )

myx!1("r ) myy

!1("r ) myz!1("r )

mzx!1("r ) mzy

!1("r ) mzz

!1("r )

#

$%%%

&

'(((

"x"y"z

#

$

%%%

&

'

(((+V("r )

)

*

+++

,

-

..

./v,n ("r ) = Ev,n/v,n("r )

Mv,T

!1=

"T

!1 Mv

!1"T with M

v

!1=

m1!1

(!

r ) 0 0

0 m2!1

(!

r ) 0

0 0 m3!1

(!

r )

#

$

%%%

&

'

(((

Rotation matrix*J.S. de Sousa et al., APL 82, 2685 (2003)J.P. Leburton, IWSG-2009, IITB, India


12/28


13/28

Energy Spectra: Effective Mass Anisotropy

Spherical nanocrystal

D = 10 nm

1 /miso =2 / (3m

t) +1 / (3m

l)



14/28

Energy Spectra: Size and Shape Effects *

Spherical Quantum Dots Truncated Nanocrystals

! Degeneracy among energy valleys

! Orbitals orientation follow the rotation of

the effective mass tensor

! Lifting of energy valleys degeneracy

! Accidental degeneracies



15/28

Crystallographic Orientation Effects*

! Different crystalline orientations are

responsible for accidental degeneracy

! !En< kBT (room temperature) for the [010]orientation

! Minibands appear for the [110] orientation

! Despite of the non-symetrical shape, energy

valleys degeneracy is recovered for the

[111] orientationJ.P. Leburton, IWSG-2009, IITB, India


16/28


17/28


18/28

Data Operation Modeling: Dynamics*

Data programming

Data erase and retention

groundstate

Bardeen Hamiltonian approach

J.P. Leburton, IWSG-2009, IITB, India *J. S. de Sousa et al, J. Appl. Phys. 92, 6182 (2002)*J. S. de Sousa et al, Appl. Phys. Lett. 82, 2685 (2003)


19/28

Charging Time Dynamics*

!Practical programming times ("100 ns) are only achieved by combining very thinoxide barriers ("20) and VG>2.0V (consistent with experiment)!Correlation between the average charging time and the number of electrons in thechannel

*J. S. de Sousa et al, J. Appl. Phys. 92, 6182 (2002)

Tunneling barrierthickness

D=7nm



20/28

*

1st consequence: redistribution of the

electrostatic potential (EP) across thedevice

EP drop in the oxide layer islarger forSiO2than for HfO2

Smaller HfO2!ECmay favor FNtunneling through the gate

compromising data write and retentionfor VG>2.5V. Thus, TCmust be

increased (>20nm)

Concerns on the dielectric

breakdown: F(HfO2)=10MV/cm

and F(SiO2)=20 MV/cm. Quality of

the oxide becomes crucial !!

High-K Oxides: Electrostatics



21/28

High-k materials increases writeperformance, but also decreaseretention time (device reliability).

Strategy: increase tunneling oxidethickness !

Main advantage: we can increase

the tunneling oxide and still obtaingood performances because of the

smaller !EC!

High-K Oxides: Programming



22/28

10 YearsSphere

3 MonthsTrunc. Sphere

11 DaysHemisphere

Retention TimeShapes

D= 70 TOX= 35

A. Thean et al., Proc. Nonvolatile Memory Technology Symp., 2000, pp.1621.

VG= 2.0V TOX= 15

The faster data are written ...

... the faster they are lost !

J. S. de Sousa et al, J. Appl. Phys. 92, 6182 (2002)J. S. de Sousa et al, Appl. Phys. Lett. 82, 2685 (2003)

A tough problem: Data retention



23/28


24/28


25/28


26/28


27/28

Optical Programming*

*J. S. de Sousa et al., Appl. Phys. Lett. 92, 103508 (2008)J.P. Leburton, IWSG-2009, IITB, India


28/28

Documents

Jpl Iitb.pres2