• Keine Ergebnisse gefunden

Nanoelectronics (unvollständig)

N/A
N/A
Protected

Academic year: 2022

Aktie "Nanoelectronics (unvollständig)"

Copied!
1
0
0

Wird geladen.... (Jetzt Volltext ansehen)

Volltext

(1)

4 ei

* kann Spuren von Katzen enthalten nicht für Humorallergiker geeignet alle Angaben ohne Gewehr *

Nanoelectronics

HINWEIS:Die Formelsammlung ist eine einfache Mitschrift, sehr ungeordnet und kann grobe Fehler enthalten. Sie dient lediglich als

Uberblick zum Fach. Wenn jemand die FS erg¨¨ anzen/¨uberarbeiten m¨ochte, einfach melden

Wichtige Begriffe:

dispersion Verteilung lattice Kristallgitter impurities Fremdstoffe to scatter streuen

Abk¨urzungen:

CVD chemical vapour deposition CNT carbon nanotube DoS Density of States

HOPG Highly Ordered Pyrolytic Graphite PMMA Polymethylmethacrylat (Acrylglas) STM Scanning tunneling microscope

1. Moores Law – scaling

1. Transistormaterial: Germanium

Transistor scaling 22nm between drain and source of a MOSFET scaling cant continue indefinitely Against Moores Law: the rising costs of fabrication, the limits of lithography, and the size of the transistor. Advan- tages of scaling: smaller, cheaper, and faster and to consume less power.

2. Quantum mechanics

Klassische Bewegungsgleichung:md2r(t) dt2 =F Classical wave equation:

1 c2

2Ψ

∂t2 − n X

i=1

2Ψ

∂x2i = 0 c=λf

ω= 2πf k=2π

λ Waves

behave as particles. Electrons and photons are both, particles and waves.

Electron Orbits:mvr=n~ nλ= 2πr Bohr atom model: En≈ −13.6Z2

n2eV Z: count of protons De Broglie Wavelength:h=pλ p=~k h= 2π~ Uncertainty principle: ∆x·∆px≥~

2 ∆E·∆t≥~ 2 transistor dimension:Lcrit≈q h

2m∗Eb

≈4 nm

withm≈0.19m0 Eb= 0.5eV

2.1. Schroedinger Equation

−~2 2m

2+V(r, t)

!

Ψ(r, t) = i~

∂tΨ(r, t) (1) Potential energyV(r, t)∈R (for HydrogenatomV(r) =−er2) HamiltonianHˆ=

−~2

2m∇2+V(r, t)

ProbabilitydensityP(r, t) = Ψ(r, t)·Ψ(r, t) =|Ψ2| Normalized:´

|Ψ(r)|2dr= 1

2.1.1 time-independent Schroedinger equation ifV(r, t)is time-independent:

Ψ(r, t) = Ψ(r) expiEt

~

⇒ HΨ(r) =ˆ EΨ(r)

1D Confinement (infinite Quantum Well):

Ψn(x) =q 2

Lxsin2mE

~ x En=2m~ k2n=~22mLπ2n2 2D ConfinementΨ(x, y) =q 4

LxLysin(kxx)·sin(kyy) En=2m~ (kx2k2y)

δ-D Confinement wirh

i=x, y, ..., δ Ψ(r) =q

QLi δ Q i=1

sin(ki·i) En=2m~ (kx2k2y)

Analytical solutions are only possible for the infinite quantum well

2.2. Quantenphysik

EP h=f·h=~·ω=hcλ λ·p=h p=~k=hkλ ~=h k=λ

2.3. Phonons

are quasiparticles to describe modes of vibrations of elastic structures of interacting particles. there are acoustic and optical phonons.

3. Semiconductors

3.1. bandstructure

FermienergieFE: H¨ochste Energie eines Elektrons beiT = 0KIsola- tor: große Bandl¨uckeEG >3eV Halbleiter: kleine Bandl¨ucke1eV <

EG<3eV kann durch thermische Energie ¨uberwunden werden Materials in columns:

IV: Si,Ge, III-V(GaAs, InP, GaN(BluRay), InSB),II-VI(CdSe, CsTe) IV- VI(PbS,PbSe)

Silicon in crystal structure: 5 per Cube

Chemical band structure: energylevels of diffrent atoms moving close together

At finite temperature some electrons can move around. n ∝ exp(T bgap)

At300K:n= 1.5×1010cm3

doping with donors(P,As) or acceptors(B,In) to lower the energy for emission or capture an electron

atoms:1023per cm3, dopants:1015per cm3 Ekin=2mp2 = ~2m2k2 d2E

dp2 =m1 Effektiv mass:m1

ef f

=1

d2E dk2 Resistorequation:RM atM atwtl conductivity:σ=1ρ =qµnni resistivity:ρ

uncrtainity for electron:∆x≥0.5·10∆v−4 vsatfor Si:2·107cms

IDS= 12µCoxW

L(VDD−VT)2 P=VDD2 Coxfmax

4. Transistors

ID= 12µnCox0 WL ·(VGS−Vth)2 µn≈250·10−4mV s2p≈100·10−4mV s2 Pcap01f CoxVDD2

fmax= Isat VDD Cox

4.1. scaling with factor

S <1 reduce areaA=W·L A0=A·S2 increase speedτ= Lv τ0=τ·S reduce powerP=V Iτ P0=P·S3

Transistorscaling in nm: 90(2003), 65(2005), 45(2007), 32(2009), 22(2011)

4.1.1 Problem of scaling 1. Tunneling across the oxide 2. Need for new lithographic techniques 3. Parasitic effects due to inteconnects 4. Melting interconnects due to voids 5. High field and breakdown effects

⇒new materials, processes and technologies needed!

High-K Material (high dielectricε) as Gate isolator: ⇒1,6·CG, 0.01·Ileak

Example Intels 45nm MOSFET: High-K with silicon gaten: Problems: un- even

Form: Normalgate, Dualgate, Trigate

Best: Surrounding Gate: CNT – high-K – metal-gate

4.2. Silicon Nanowire

Fabrication: growth on a gold(Au) particle

4.3. GaN - Transistors

Why GaN?

• Wide bandgaps of GaN and AlGaN (high breakdown volt.)

• high drift velocity(hf)

• strong piezoelectric effekt

• High temperature operation

HEMT-Transistor: Two substrate materials, doped and undoped

⇒electrons move on a 2D-Sheet Cut-Off-FrquencyfT= vsat

2πLg:gm= 1(no amplification anymore) Oxide-CapacitanceCox= εtox0εr

T-Gate: smooth electric field in the channel.

4.4. Quantum Wire

Ideal: just one subband in two dimensions But for good conductan- ce(mobility, drift velocity) one need20nmFabrication Methods: Stressor, Etching, Ion implantation, Vicinal Growth

Split Gate Transistor: 1D tunnel in the gate between source and drain.

electron wave transistor:

5. Graphene

2D Network of 3D Carbon Atoms.

Stacked Layers of Graphene form Graphite.

applications: seperation membranes, capacitors

a

1

a

2

y x

b d

b≈0.14 nm d= 2b≈2.8 nm a1= 32bx+

√ 3 2 by=

√ 3 2 ax+12ay a2= 32bx−

√ 3 2 by=

√ 3 2 ax−12ay a1

= a2

=a=√

3b≈0.246 nm kxk=

y

= 3b

5.1. Properties

thinnest material sheet imagineable extremly strong (5 times stronger than steel)

semimetall: better conduction than metal, can switched ON and OFF very light, good head conductor size of one cell: edged≈0.14nm, edge2edgea=√

3d

5.2. production

•Exfoliated Graphene: peeling HOPG with foil. Good for scienece not for manufacturing

•Epitaxial growth: silicon carbide(SiC) is heated (>1100C) to re- duce it to graphene.

5.3. Carbon-Nano-Tubes CNT

Propertys: diameter:d≈10 nm

Application: wires, transitors, sensors, Molecular tweezers Single Walled and Multiwalled: SWCNT: single layer of graphite(graphene) rolled up as cylinder

Kind of curls: zig-zag(n,0), armchair(n, n), quiral(n, m) chiral vector (tube circumfence):Ch=na1+ma2

translation vectorT= [(2m+n)a1−(2n+m)a2]/gcd(n, m) Tube-Diameter: dT = kchk

π = aπp

n2+nm+m2 with a= 0.246nm

Kind of Nanotube:

(metalic if(n−m)/3∈N0 semiconductor else

Bandenergy in dependency ofk:

E(k) =ε0±t s

1 + 4 cos √

3akx 2

cosaky

2

+ 4 cos2aky 2

Periodic Boundary Conditions:C>h·k= 2πn

Homepage: www.latex4ei.de – Fehler bittesofortmelden. von LaTeX4EI - Mail:info@latex4ei.de Stand: 7. Juli 2014 um 15:16 Uhr 1

Referenzen

ÄHNLICHE DOKUMENTE

21 Crisis Group interviews, Metin Çorabatır, founding member, Research Center on Asylum and Mi- gration (IGAM), Ankara, December 2013, international refugee official,

The discussion of a universal low-energy scale for specific model parameters away from the abstract true Kondo scaling limit with the bandwidth by far the largest energy is important

The paradigm of Technical Inheritance allows monitoring the manufacturing and usage of a component, to analyse and employ the collected data into the development process of

As the startup grows, the complexity of the software system and of the or- ganization eventually reaches a point where a flat organizational structure and a monolithic

For low eective packing fractions and suciently large shear rates respectively frequencies, such that no crystallization occur in the sample, the ow curves and the corresponding

The main lessons that were drawn from this analysis are summarized as follows: (i) small scale, less capital-intensive, and short lifetime technologies can diffuse faster (shorter

Several phenomenological descriptions, such as the von Bertalanffy growth model, have been widely used to describe size-at-age and individual growth across a diverse range of

We introduce S INDY , an algorithm that efficiently discovers all unary inclusion dependencies of a given relational dataset in a distributed fashion and that is not tied to main