Title: RELEVANCE OF DEVICE MODELING AND NUMERICAL SIMULATION FOR :
1RELEVANCE OF DEVICE MODELING AND NUMERICAL
SIMULATION FOR
- INTERCONNECT SIMULATION
- ELECTROMAGNETIC WAVE EFFECTS ON MICROWAVE
TRANSISTORS - FROM THE ANALYSIS OF SOME RELEVANT LITERATURE
- ADVANCED DEVICE MODELING (HIGHER ORDER MOMENT
EQUATIONS) AND NUMERICAL METHODS (HIGH
RESOLUTION CONSERVATIVE SCHEMES) REQUIRED !
2MODELS INCORPORTATED IN COMMERCIAL SIMULATORS
- ISE
- DRIFT-DIFFUSION
- ENERGY TRANSPORT
- SIMPLIFIED HYDRODYNAMICAL
- THERMAL
- PARAMETERS PHENOMENOLOGICALLY ADJUSTED
- LACKING
- PHYSICS BASED MODELS
- ELECTROMAGNETIC INTERACTIONS
- INTERCONNECTS
3(No Transcript)
4MODELLING AND SIMULATING INTERCONNECTS
- ON CHIP INTERCONNECTS ARE A LIMITING FACTOR TO
THE OVERALL PERFORMANCE OF CIRCUITS. ON CHIP
WIRING DELAYS ARE A SIGNIFICANT PORTION OF THE
TOTAL CHIP DELAY TIME. - METAL-INSULATOR-SEMICONDUCTOR INTERCONNECTS ARE
OF PARAMOUNT INTEREST - ELECTROMAGNETIC FIELD COUPLING WITH DEVICE CHARGE
CARRIERES TRANSPORT IS REQUIRED FOR ACCURATE
DESCRIPTION - G.WANG, R.W.DUTTON, C.S..RAFFERTY, DEVICE LEVEL
SIMULATION OF WAVE PROPAGATION ALONG
METAL-INSULATOR SEMICONDUCTOR INTERCONNECTS, IEEE
TRANS ON MICROWAVE THEORY AND TECHNIQUES, VOL.50,
(2002), pp.1127-1137.
5(No Transcript)
6(No Transcript)
7(No Transcript)
8MAIN RESULTS
- DETERMINATION OF THE SLOW SPEED FOR THE
FUNDAMENTAL TM MODE - AT VARIANCE WITH THE USUAL TREATMENT WHERE THE
SEMICONDUCTOR IS TREATED AS A MEDIUM WITH UNIFORM
CONDUCTIVITY, THE DEVICE DESCRIPTION ENABLES TO
I - ANALYZE THE HIGHER MODES GENERATION THROUGH THE
NON LINEAR BEHAVIOUR OF THE SEMICONDUCTOR - TREAT BOTH SMALL SIGNAL AND LARGE SIGNAL CASES
9FUNDAMENTAL MODE PROPAGATION Ex AND Ez
10FUNDAMENTAL MODE PROPAGATION ELECTRON AND HOLE
CONCENTRATIONS
11SECOND HARMONIC Ex AND n
12HIERARCHY OF MOMENT EQUATIONS
- THE NON STATIONARY DRIFT DIFFUSION EQUATIONS ARE
THE FIRST TERM IN THE HIERARCHY OF MOMENT
EQUATIONS (1) AND (3)
13RELATIVE MAGNITUDES OF THE RELAXATION TIMES
- IN THE HIERARCHY OF THE MOMENT EQUATIONS THE
RELAXATION TIMES FOR MOMENTUM, ENERGY, ENERGY
FLUX, FLUX OF ENERGY FLUX, ETC. APPEAR - IT IS POSSIBLE TO LIMIT TO THE DRIFT-DIFFUSION
EQUATIONS ONLY IF THE OTHER RELAXATION TIMES
COULD BE CONSIDERED NEGLIGIBLE - HOWEVER
- MONTE CARLO SIMULATIONS (MUSCATO) SHOW THAT THE
RELAXATION TIME FOR ENERGY IS LARGER THAN THE
OTHERS AND THE MOMENTUM AND ENERGY FLUX
RELAXATION TIMES ARE COMPARABLE !
14(No Transcript)
15(No Transcript)
16EXTENDED MODELS
- INCORPORATE BALANCE EQUATIONS FOR MOMENTUM,
ENERGY AND ENERGY FLUX - THE RELAXATION TIMES FOR MOMENTUM AND ENERGY FLUX
ARE OF THE SAME ORDER OF MAGNITUDE - THE ONSAGER RECIPROCITY THEOREM OF IRREVERSIBLE
STATISTICAL PHYSICS MANDATES THAT AT LEAST THE
ENERGY FLUX BALANCE EQUATIONS MUST BE INCLUDED
(MUSCATO, 1998) - THE PARAMETERS APPEARING IN THE PDE MUST BE
OBTAINED FROM THE PHYSICAL MODEL, EITHER THROUGH
MONTE CARLO DATA FITTING OR BY ASYMPTOTIC
SOLUTIONS OF THE BOLTZMANN EQUATION.
17ELECTROMAGNETIC WAVE EFFETCS ON MICROWAVE
TRANSISTORS
- WHEN SEMICONDUCTOR DEVICES ARE OPERATED AT
SUFFICIENTLY HIGH FREQUENCY THE QUASI STATIC
(POISSON BASED) SEMICONDUCTOR DEVICE MODELS FAIL
TO REPRESENT ACCURATELY THE COUPLING OF THE
ELECTROMAGNETIC WAVE PROPAGATING IN THE DEVICE
WITH THE CARRIERS. - THE FULL MAXWELL EQUATIONS MUST BE COUPLED TO THE
DEVICE EQUATIONS (MOMENTS EQUATIONS) - ALSUNAIDI, IMTIAZ, EL-GHALAZY, ELECTROMAGNETIC
WAVE EFFECTS ON MICROWAVE TRANSISTORS USING A
FULL WAVE TIME DOMAIN MODEL, IEEE TRANS ON
MICROWAVE THEORY AND TECHNIQUES, 44, (1996),
799-899
18(No Transcript)
19DEVICE MODEL USED BY ALSUNAIDI ET AL.
20SOME RESULTS
21(No Transcript)
22LIMITATIONS
- ELECTROMAGNETIC PROPAGATION THROUGH THE DEVICE
- THE RELAXATION TIMES OF THE ENERGY FLUX ARE OF
THE SAME ORDER AS THOSE FOR MOMENTUM. - THEREFORE THE ENERGY FLUX MOMENT EQUATION SHOULD
BE CONSIDERED - THE DEVICE MODEL, AS IN THE PREVIOUS CASE, DOES
NOT OBEY THE ONSAGER RECIPROCITY THEOREM.TO
RESTORE THE ONSAGER RECIPROCITY THEOREM AT LEAST
THE ENERGY FLUX BALANCE EQUATION MUST BE INCLUDED - NECESSITY OF EXTENDED MODELS (HIGHER ORDER
MOMENTS).
23(No Transcript)
24EXTENDED HYDRODYNAMICAL MODELS
25(No Transcript)
26(No Transcript)
27(No Transcript)
28(No Transcript)
29(No Transcript)
30APPLICATION OF THE METHOD
31(No Transcript)
32(No Transcript)
33(No Transcript)
34TEST 1D DIODE N N N
35TEST FOR THE EXTENDED MODEL WITH 1D
STRUCTURESMUSCATO ROMANO, 2001
36(No Transcript)
37(No Transcript)
38(No Transcript)
39(No Transcript)
40(No Transcript)
41(No Transcript)
42TEST OF THE EXTENDED (MEP-CLOSURE) MODELS ON A 2D
BIPOLAR TRANSISTOR
43ENERGY BAND AND HOT ELECTRONS
44MOMENTUM PRODUCTION (y-COMPONENT)
45(No Transcript)