24 July 2002 Work In Progress – Not for Publication Modeling and Simulation ITWG Jürgen Lorenz - FhG-IISB ITWG/TWG Members H. Jaouen, STM * W. Molzer,

24 July 2002 Work In Progress – Not for Publication

Modeling and Simulation ITWG

Jürgen Lorenz - FhG-IISB

ITWG/TWG MembersH. Jaouen, STM*

W. Molzer, Infineon*

R. Woltjer, Philips*

G. Le Carval, LETIJ. Lorenz, Fraunhofer IIS-B*

W. Schoenmaker, IMEC* supported by EC User Group UPPER+

T. Wada, ToshibaK. Nishi, SELETE Japanese TWG 16 industrial members

C. Riccobene, AMDL. Richardson, HPM. Giles, INTELM. Orlowski, MotorolaM. Meyyappan, NASAV. Bakshi, SEMATECHE. Hall, ex-chairperson/Motorola

J.-H. Choi, HynixK.H. Lee, Samsung

S.-C. Wong, TSMC


Key Messages

Update of key messages from 2001 ITRS:

• Technology modeling and simulation is one of a few enabling methodologies that can accelerate development times and reduce development costs: Assessment 25% in 2001, 35% in 2003, 40% in 2006

• Strong cross-cut links to the other ITRS sections were established - major goal of ITWG activities to further extend these

• Accurate technology experimental characterization is essential.

• Modeling and simulation provides an ‘embodiment of knowledge and understanding’. It is a tool for technology/device optimization and also for training/education.


Equipment related•Equipment/feature scale•Lithography

Feature scale•Front End, Back End•Device

IC-scale•Circuit elements•Package modeling•Interconnect performance modeling

•Materials Modeling •Numerical Methods

Technology Modeling SCOPE & SCALES

(Chapter sub-sections in blue)


Trends / changes from 1999/2000 edition of ITRS

• Increased need for fundamental materials modeling and relating those results to electronic properties ( e.g. gate stack )

• Need much better techniques / methodologies for exploring “end of the roadmap” issues.

• Stronger need for RF simulation methodologies.

• Need greater tie between models and chip design methodologies

• Need better analytical and characterization techniques to aid in the development of predictive models.

• Relevance of advanced numerical methods and algorithms increasing & more detailed in roadmap

Key Messages ( con’t)


Trends / changes from 2001 edition of ITRS• New long-term challenge “Compact modeling including

statistics”• Adapt “summary of issues” in challenges list to current

technical progress (e.g. skip 248 nm)• Adapt details of near-term requirements to current technical

progress:- esp. lithography status and roadmap- esp. multi-level hierarchical simulation

ITWG actions in 2002• Update 2001 tables• Further increase interactions with other ITWGs & impact of

simulation

Key Messages ( con’t)


Difficult Challenges > 65 nm


Needs• Efficient simulation of full-chip interconnect

delay

• Accurate 3D interconnect model; inductance, transmission line models

• High-frequency circuit models that including

– non-quasi-static effects

– predictive noise behavior

– coupling

• Predictive, scalable inductor model

• Reduction of high-frequency measurements needed for parameter extraction for active and passive devices

Difficult Challenges

High-Frequency Circuit Modeling (>5Ghz)

gate

g2

bulk

drain sources1 d1s2 s3 s4 s10 s9s8

No series resistanceNo DIBL,No static feedbackNo overlap capacitance

g1

R gate

R bulk, central

R bulk, drain R bulk, source

C jun,s C jun,d

b1

C gso C gdo

R source R drain

+

_ V

T1

(From Philips)



Modeling of Ultra Shallow Dopant Distributions (Junctions, Activation), and Silicidation

Needs

• Dopant models & parameters (damage, high-concentration, activation, metastable effects, diffusion, interface and silicide effects)

• Characterization tools for ultra-shallow geometries and dopant levels

Source: A. Claverie, CEMES/CNRS, Toulouse, France



Modeling of Deposition and Etch Variations, and Feature Variations Across a Wafer

Needs

• Fundamental physical data ( e.g. rate constants, cross sections, surface chemistry).

• Reduced models for complex chemistries

• Linked equipment/feature scale models

• CMP (full wafer and chip level, pattern dependent effects)

• Next generation equipment /wafer models (e.g. 450 mm wafers )

Simulated across-wafer variation of feature profile for a sputter-deposited barrier.



Modeling of Lithography TechnologyNeeds

• Predictive resist models (incl. mechanical stability)

• Resolution enhancement techniques; mask synthesis (OPC, PSM)

• 193 nm versus 157 nm evaluation and tradeoff methodologies.

• Next generation lithography system models: EUV, electron projection, maskless lithography

Printing of defect on phase-shift mask: bump defect (top) vs. etch defect (lower)



Gate Stack Models for Ultra-thin Dielectrics

Needs

•Electrical (breakdown, transport, reliability) & processing models for alternative gate materials and dielectrics

•Be able to model dielectric constant, surface states, defects, band gap, .... from process/material conditions.

MSI

- Band structure- Carrier effective mass ...

Direct Applications- Defect states- Leakage current- Impurity transport- Processing recommendations ...

Use atomistic models to predictphysical and electronic propertiesof materials; eg, HfO2.

Potentials in a thinSiO2 layer.

-1.5

-1

-0.5

0

0.5

1

-10 0 10 20 30 40 50 60

En

erg

y (

eV

)

Distance (nm)

25 nm MIT MOSFET Density of Statesand IV Curve

VDS=1.2V

10-11

10-10

10-9

10-8

10-7

10-6

10-5

10-4

10-3

0 0.2 0.4 0.6 0.8 1

MIT 25nm: Quantum Dissipative vs Ballistic ID-V

GS

Quantum BallisticQuantum (=12.5meV)

I D (

A/

m)

VGS

(V)

VDS

=1.2V


Difficult Challenges < 65 nm


Needs

• Stress voiding

• Electromigration

• Piezoelectric effects

• Textures

• Fracture ( thin film and bulk)

• Adhesion

• Thermo-mechanical Stress in a 80x4 m2 bamboo segmentdue to electromigration


Thermo-Mechanical modeling

(IMEC)


Other 2001 ITWG RecommendationsGeneral

• Support increased cross-discipline efforts that bring in experts from physics, chemistry, mathematics, and other fields to aid in solving these difficult challenges

• Need adequate research funds for universities and laboratories for directed long range research

• Explore ways of ‘standardizing’ and / or opening up some of the universally used modeling and simulation modules so the focus is on value-add efforts

• Need improved methodologies for evaluating the impact of modeling and simulation

• Need a hierarchy of software tools - spread sheets to ab-initio


Other 2001 ITWG Recommendations

• Equipment Modeling

• Equipment suppliers should supply physical models and modeling information with equipment

• Process/Device Modeling

• Need continued modeling effort on improving process simulators

• Reliability models for circuit design and technology development


Other 2001 ITWG Recommendations• Package Simulation

• Need co-design integration software tools- Thermal, mechanical, and electrical- Chip and package- RF capability

• Circuit Modeling

• Increased effort on industry standard circuit models

• Better methodologies for linking process and new device effects to designers

• Ioff/Ion software tools - e.g leakage prediction


New 2002 ITWG Recommendations

• Tables modified in details

• New long-term challenge “Compact Modeling Including Statistics”

• Long-term requirement for emerging devices: “Nanoscale simulation capability including accurate quantum models”

• Long-term requirement for Package Modeling - electrical/optical models: “Reliability prediction in coupled modeling”


Thank You

Documents

24 July 2002 Work In Progress – Not for Publication Modeling and Simulation ITWG Jürgen Lorenz - FhG-IISB ITWG/TWG Members H. Jaouen, STM * W. Molzer,