p Oled Materials Device Operation

Polymer OLED Materials and OLED100 Summer School, 2011

D Al d D t

Polymer OLED Materials and Device Operation

Dr Alexander Doust

Cambridge Display Technology Limited, g p y gy ,Building 2020, Cambourne Business Park, Cambourne, Cambridgeshire, CB23 6DW

OutlineOutline

Materials and Material PropertiesMaterials and Material PropertiesMaterials and Material PropertiesMaterials and Material Properties

Device OperationDevice Operation

Improving Device PerformanceImproving Device Performance

pp

ConclusionsConclusions

CDT 2011 www.cdtltd.co.uk (Company Number 02672530)2

Introduction to CDTIntroduction to CDT

CDT founded in 1992 to exploit Polymer-OLED Lighting ltechnology technology for display applications

CDT is now owned by Sumitomo Chemical Group

panels

Group159 permanent staff

105 scientific & technical employees with degree Indicators56% (59) PhD

Multidiciplinary scientific environment:Analytical/Organic/Inorganic/Physical/Theoretical Analytical/Organic/Inorganic/Physical/Theoretical ChemistryMaterials Science & Polymer ChemistryElectronics & Engineering

Flat panel displaysElectronics & EngineeringDevice Physics, Optics, ModellingMicroscopy

CDT 2011 www.cdtltd.co.uk (Company Number 02672530)3 3

Design (3D-CAD)Formulation

OutlineOutline




pp



Overview OLED materialsOverview - OLED materials

Small MoleculeSmall Molecule OLEDs (SM-OLEDs)

O

NAl

N

O

Invented 1985by Tang, van Sl k (K d k)

NN

O

Alq3

N N

Biphen

Polymer OLEDs (P OLEDs)

Slyke (Kodak) Alq3

(P-OLEDs)

Invented 1989 by Burroughes

nn

Burroughes, Friend, and Bradley (Cambridge)

PPV Poly(9,9-dioctylfluorene)


(Cambridge)

Materials - examplesMaterials - examples

Small Molecules Polymers

On

O

Me O n

PP VNN

TPD

n

MEH-PPVNN

NBPO

N

PP P nNN

O

AlO

Poly(9,9'-dioctlyfluorene)Alq3 N N

Biphen


Molecular Orbitals & Energy LevelsMolecular Orbitals & Energy Levels

Vacuum level

Higher unoccupied levels

LUMO (lowest unoccupied molecular orbital)

Eg (Bandgap) determines colour of emitted lightEnergy

Lower occupied levels

HOMO (highest occupied molecular orbital)

Number of levels, energies, spacing, etc. determined by molecule ie. # and type of atoms.

Most important orbitalsfor OLED device physics

by molecule ie. # and type of atoms. Number of electrons directly linked to molecular

charge.0, 1 or 2 electrons per level (opposite spins in case of


physics , p ( pp p2 electrons).

Energy is relative to vacuum level

Charge Carriers & Organic Semiconductor Band Diagram

Vacuum level

LUMO levelLUMO level

y

e-e-

o

n

e

n

e

r

g

y

HOMO levelE

l

e

c

t

r

o h+h+

E itNeutral Hole Electron Excitonlocalised on one molecule

Neutral Molecule

Hole (cation-radical)

Electron(anion-radical)


Charge TransportCharge Transport

Hopping transport Localized carrierspp g p

charges hop from site to site along or b t l h i

Localized carriers Nuclei relax around charge

between polymer chains

there is a barrier to hopping between sites

-

there is a barrier to hopping between sites

hopping probability depends on AN

CA

temperature, electric field, molecule separation and level of disorder in the film

NODE

THOD

charges drift (under the influence of the electric field) and diffuse (as a result of d it di t )

EDE


density gradients)

2 types of excitons are possible2 types of excitons are possible

TRIPLETSINGLET

The spin of the LUMO electron can pair with the spin of the HOMO

TRIPLETThe spin of the LUMO electron does not pair with the spin of the HOMO l tspin of the HOMO

electron.

Can recombine to

electron.

Can only recombine to emit light in a

UDCan recombine to emit light in any material (fast).

to emit light in a phosphorescent material (slow).

Electrons and holes recombine forming either a triplet or singlet excitonSinglets can decay to emit light in fluorescent and phosphorescent g y g p pmaterials.Triplets can only decay to emit light in phosphorescent materials (lower energy) Non-radiative (thermal) decay will be present in any material.


energy) Non radiative (thermal) decay will be present in any material.Spin statistics suggests a 3:1 ratio of triplets:singlets - Not necessarily the case for all materials.

Organic Semiconducting MaterialsOrganic Semiconducting Materials

What makes an organic material semiconducting?g gThere are two basic types of covalent bond:

Sigma () bond Pi () bond

C C C C

StrongTi htl b d

WeakL l b dTightly bound

LocalisedLoosely boundDe-localised


Semiconducting Molecules and PolymersSemiconducting Molecules and Polymers

As the de-localised pi-orbitals form a bond they split into bonding (HOMO) l l d ti b di (LUMO) l l ith i d tenergy levels and anti-bonding (LUMO) energy levels with semiconductor

properties emerging as the delocalisation extends.

Conduction Anti-BondingOrbital Conduction

BandOrbital

Bonding LUMO

4t1LUMO wave function

Orbital HOMOValence Band 4t1

Degeneratepi levels

Non-degenerate molecular orbital levels

Quasi-continuousenergy bands resulting from overlap of many pi

HOMO wave function

Molecular orbitals of


orbital levels overlap of many pi orbitals poly(para-phenylene)

(PPP)

Phosphorescent materialsPhosphorescent materials

In order for triplet excitons to emit spin orbit coupling isIn order for triplet excitons to emit, spin-orbit coupling is required.

Spin-orbit coupling is the interaction between the p p gmagnetic moments that arise from the spin and the orbital angular moments of an electron.Thi ff t i t i h tThis effect is strong in heavy atoms.

Transition metals are common candidates to enable phosphorescence from organic moleculesphosphorescence from organic molecules.

N

Ir

3


Iridium tris(phenylpyridine) aka. Ir(ppy)3

Phosphorescent dopantsPhosphorescent dopants

Example Iridium-based phosphorescent dopants for RGBExample Iridium-based phosphorescent dopants for RGB emittersTo use phosphorescent materials, their exciton energy must be lower than the host triplet.

e.g. a red phosphorescent emitter needs a blue hostS1

S1

SIr

O

T1T1

S1

N2

IrO

F

NIr

3N

2

FIr

N

O O

HOST EMITTER

S0 S0


2

OutlineOutline




pp



Light Emission from Organic PolymersLight Emission from Organic Polymers

Light emission results from recombination of injected chargesCharges can be introduced optically by light absorption

Electron injection LUMO

Holeinjection

LightHOMO

injection

Charge injection from contacts

RecombinationExciton formation

Exciton decayLuminescencefrom contacts Exciton formation

(neutral excited state)

Luminescence


Excited States in PLEDsExcited States in PLEDs

p+ p-Fl Phosphorescence

recombinatio

p

S S

Fluorescence PhosphorescenceIf Emissive

tripletsinglet

n

( )1 ( )+1

Spin = 0 Spin = 1

3 Statestriplet exciton

singlet exciton

radiativradiativ nonnonnon radiativenon radiative

( )2 ( )+2

3 States

radiative decayradiative decay

non-radiative decay

non-radiative decay

non-radiative decaynon-radiative decay

Measure generation and decay of singlets, triplets and charges to g y g p gdetermine singlet:triplet ratioIF formation probability of singlet and triplet states is identical then expect singlet:triplet ratio to be 1:3


expect singlet:triplet ratio to be 1:3 It is possible to generate a higher ratio of singlet states

Completed OLED structureCompleted OLED structureAnode: Large barrier for electron

injectionCathode: Small barrier for electron

injection

Vacuum level

injection(electrode and LUMO levels

mismatched)

injection(electrode and LUMO levels are aligned)

LUMO levelFermi LevelCathodeEnergy Cathode

Fermi LevelAnode

HOMO level

HOMO-LUMO gap determined by desired wavelength ( red ~2V, blue

LEDs device function

before contact after contact

LEDs device function

V=0

fermi levels (workfunctions) align thermal equilibrium thermal equilibrium no charge inside the LEP layer no band bending


Forward bias to Flat Band VoltageForward bias to Flat Band Voltage

V = Vbi =~ f(cathode)-f(anode) Flat band Voltage = V Vbi f(cathode) f(anode) gminimum bias for charge injection

CDT 2011 www.cdtltd.co.uk (Company Number 02672530)20 20

Forward Bias light emissionForward Bias light emission

Occurs for all V > Vbi Space Charge Limited CurrentOccurs for all V > Vbi p g

cathodeV

anode

+


Problems with simple structuresProblems with simple structures

A single organic layer OLED has a number of drawbacksIf the charge recombination occurs at either electrode the emission can be quenchedthe emission can be quenched.The charges must be balanced to ensure recombination always towards the middle.yCharge balance typically changes with drive level, and often with age, so maintaining this balance is virtually impossible.


Efficiency RelationshipEfficiency Relationship

Relationship between efficiency and controllingRelationship between efficiency and controlling factors

Quantum efficiency = eff = photons outcharges flowing in circuitFraction of excitons that can

di ti l d 25% 90%

charges flowing in circuit

eff = q rst extradiatively decay 25% - 90%

Optical outcoupling20%-40% Ratio of excitons toRatio of excitons to

charge injected EQE:50% - 100% Efficiency of radiative decay of

singlet excitons (PL efficiency) depends on polymer and purity


depends on polymer and purity40 - 80%

Maximising Singlet:Triplet RatioMaximising Singlet:Triplet Ratio

For a fluorescent material, only singlet excitons are emissive.Simple quantum mechanics suggests a 1:3 S:T ratio and a 25% QE limit.Work at Cavendish Lab, Cambridge suggests ratio is ~1:1 for polymers (1:3 for small molecules)These are measurements of monomers and polymers with the same unit cell and identical techniques.

0 8

1.0

o

n

290 Kthickness dependence

Wilso

0.6

0.8

r

a

t

i

o

n

f

r

a

c

t

i

PolymerPolymers

ST Ratio~0.5

on et al, N

0.2

0.4

i

n

g

l

e

t

g

e

n

e

r

MonomersST Ratio~0.25

Nature 413

0.00 50 100 150 200 250

S

i

MonomerST Ratio 0.25 3, 828 (200A reason for the increased

S T ti ill b h


Thickness (nm)

01)S:T ratio will be shown later.

Why are polymers different from small molecules?

Small Molecules : electron-hole capture at separation 10 nmPolymers: electron and hole arrival on polymer chain, and

subsequent recombination to exciton

(iii) electron-hole capture on chainh

(i) hole t

(ii) electron

he

capture capture

The singlet is higher energy and therefore larger. This is accentuated in 1D l tpolymer systems.

This favours singlet bound state at e-h capture (typical range 10 nm) (Beljonne et al J. Chem. Phys. 102, 2042, 1995)


SM-OLED Device StructureSM-OLED Device Structure

Band structure TransportsElectron hole

Generic Structure

Blocks electrons, excitons from exciting EML

Transports electrons to emissive layer

recombination, emission

Cathode

Electron Injection layer (EIL)

Generic Structure

CathodeE

ET

Assists injection into HTL

Electron Tranport Layer (HTL)

Emissive Layer (EML)

Exciton/Hole Blocking Layer

HTL

EML

TL

Assists injection into ETL Emissive Layer (EML)

Hole Transport Layer (HTL)

Exciton/Electron Blocking Layer

Anode

Transports holes

into ETL

Hole Injection Layer (HIL)

Anode

Blocks holes, excitons from exciting EML

pto EML


Materials and device structure

Glass/Plastic substratT t d ( ITO)

Low workfunctiont lli th d

Organic Semiconductor layers

Ca G~100-150nm

Transparent anode (e.g. ITO)metallic cathode

ITO

HiLiLLEP

athode

Glass

F8 PFB

e

F8 TFBN N

0.05C8H17 C8H17 n0.95

F8 PFB

0.5

N

F8 TFB

C8H17 C8H17 n0.5

Random copolymer A B copolymer


Random copolymer F8:PFB 95:5

A-B copolymerF8:TFB 50:50

Charge MobilityCharge Mobility

h (LEP) < e (iL) < e (LEP) < h (iL)

Model materials satisfy mobility requirements for ideal RZ

Schematic and summary of optimized material mobility and thickness design rules.The model materials satisfy the mobility requirements for an idealThe model materials satisfy the mobility requirements for an ideal Recombination Zone (RZ)The RZ profile fits very well to exponential decay within LEP peaking at IL:LEP interface


at IL:LEP interface.

Device structure

e-

Ca Gme > h

h

ITO

HiLiLLEP

athode

Glass

h+

~65nm ~15nm LEP thickness and mobilities - Optimum RZ for outcoupling

h+

LEP thickness and mobilities - Optimum RZ for outcoupling iL - hole injection, efficiency and lifetime HIL and ITO thicknesses tuned for colour and outcoupling Electrodes / charge injection layers - thermally stable


Electrodes / charge injection layers thermally stable

Simulation RZ and IQE

1111110 8

1f Normalised Recombination Zone

QExciton formation efficiency

0.6

0.8

0.6

0.8

0.6

0.8

0.6

0.8

0.6

0.8

0.6

0.8

0 4

0.6

0.8

LEP iL

0

0.2

0.4

0

0.2

0.4

0

0.2

0.4

0

0.2

0.4

0

0.2

0.4

0

0.2

0.4

0

0.2

0.4

0

0 10 20 30 40 50 60

0

0 10 20 30 40 50 60

0

0 10 20 30 40 50 60

0

0 10 20 30 40 50 60

0

0 10 20 30 40 50 60

0

0 10 20 30 40 50 60

111111

0

0 1 2 3 4 5 6 7 8VVbiNormalised Current

Distance from cathode (nm)

0.6

0.8

1

0.6

0.8

1

0.6

0.8

1

0.6

0.8

1

0.6

0.8

1

0.6

0.8

1Jh

LEP iL

Photon Generation Zone (PGZ) zone gets narrower at higher fields

0.2

0.4

0.2

0.4

0.2

0.4

0.2

0.4

0.2

0.4

0.2

0.4

JeLEP iL g e e dshole rich at low voltage,

electron rich at high voltage


0

0 10 20 30 40 50 60 70

0

0 10 20 30 40 50 60 70

0

0 10 20 30 40 50 60 70

0

0 10 20 30 40 50 60 70

0

0 10 20 30 40 50 60 70

0

0 10 20 30 40 50 60 70

eDistance from cathode (nm)

Model materials Device IVLModel materials Device IVL

1000 0.05

110100

m

2 0.03

0.04

Ewithout iL

with iL with iL

0 0010.010.11

m

A

/

c

m

0 01

0.02

E

Q

Ewithout iL

without iL

0.00010.001

4 3 2 1 0 1 2 3 4 5 6 7 8 9V0.00

0.01

0 1 2 3 4 5 6 7 8 9V

5% EQE can be achieved with optimized device, at CIE-y ~0.13.In this optimized device structure, the importance of the interlayer in achieving this efficiency is clear and in agreement with the expectations from the simulations.


OutlineOutline




pp



Case Study 1: Fluorescence quench sites

1000 PL intensity

800

900

m

2

Constant current density

n

e

s

s

15000

20000

n

t

s

undriven

driven

y

600

700

800

C

d

/

m

B

r

i

g

h

t

n

0

5000

10000

C

o

u

n driven

500

600

0 1000 2000 3000 4000hrs

T50

B

400 450 500 550nm

No change in colour

0 1000 2000 3000 4000

Key challenge for P-OLED is extending lifetime.Analysis of PL of device driven to 50% of initial brightness reveals aAnalysis of PL of device driven to 50% of initial brightness reveals a 30% drop in PL intensity and is a real change in the material PLQE

PL quenching site formation is dominant degradation mechanism


Cause of PL decayCause of PL decay

Bipolar device1

Hole only device1

Electron only device 1

p

0.4

0.6

0.8

I

n

t

e

n

s

i

t

y

y

0.4

0.6

0.8

I

n

t

e

n

s

i

t

y

0.4

0.6

0.8

I

n

t

e

n

s

i

t

y

drivenundriven

drivenundriven

drivenundriven

0

0.2

400 450 500 550 600nm0

0.2

400 450 500 550 600nm0

0.2

400 450 500 550 600nm

PL quenching site formation is dominant degradation mechanism.The PL decay from single carrier devices is shown to be remarkably stable.


y g yThis strongly suggests that excitons are required to generate PL quenching sites.

PL recovery matched by EL recovery

PL decay can be split into permanent and recoverable components.

PL recovery matched by EL recovery

The recovery in the optically excited PL efficiency is matched by a recovery in the electrically driven EL efficiency as shown below:

Bake: 120C 30minsBake: 120C 30mins

PL d b lit i t t d bl t PL decay can be split into permanent and recoverable components 40-50% of EL and PL decay at T50 can be recovered by baking!! The threshold temperature for PL recovery is the LEP Tg and the recoverable component of decay can by cycled many times


recoverable component of decay can by cycled many times.

Reversible portion of degradationReversible portion of degradation

1

L/L0Bake above Tg steps

0.6

0.8L/L0

0.2

0.4

0

0 0.5 1 1.5 2 2.5 3time

The characteristic lifetime of the recoverable degradation is similar after first and second bakes suggesting that intrinsic stability of the recoverablefirst and second bakes, suggesting that intrinsic stability of the recoverable portion of degradation indeed reverts back to its pristine condition and not to some weakened intermediate state.I d LEP t i l t bl t it h th t thImproved LEP materials are more stable to excitons, such that the recoverable PL dominates, with >95% PL recovery.Tackling the recoverable PL quenching sites is the key to P-OLED stability.


Case study 2 loss of triplet yieldCase study 2 loss of triplet yield

1

Fast decay Model materials

0 1

i

s

e

d

)

~24ns (Device RC limited)

No triplet quencher

0.1

t

y

(

n

o

r

m

a

l

Residual Delayed EL

0.01

E

L

i

n

t

e

n

s

i

t Delayed EL unchanged by reverse bias pulse

-10V reverse

0.0 5 0x10-7 1 0x10-6 1 5x10-6 2 0x10-6 2 5x10-61E-3

E

pulsereverse bias applied

0.0 5.0x10 1.0x10 1.5x10 2.0x10 2.5x10Time /s

~20% of luminescence has a very long lifetime (msecs)


Delayed electroluminescence does not originate from trapped charges

Delayed ElectroluminescenceDelayed Electroluminescence

Model materials

No triplet quencherTriplet Density1

m

a

l

i

s

e

d

)

No triplet quencher

(Triplet Density)20.1

8

0

n

m

)

(

n

o

r

m

0.01

y

/

d

T

/

T

(

7

8

6 6 6 61E-3

E

L

i

n

t

e

n

s

i

t

y

Delayed EL

0.0 5.0x10-7 1.0x10-6 1.5x10-6 2.0x10-6 2.5x10-6E

Time /s

O i i f d l d fl i TTA T T S S CDT 2011 www.cdtltd.co.uk (Company Number 02672530)38

Origin of delayed fluorescence is TTA : T1 + T1 S1 + S0

Driving Effect on Triplet DensityDriving Effect on Triplet Density

Model materials, No triplet quencher

1.2x10-4

/

T

)

20000

25000 Pristine Driven to T50 Triplets

quenched by

4.0x10-5

8.0x10-5

t

D

e

n

s

i

t

y

(

d

T

/

5000

10000

15000

P

L

C

o

u

n

t

s Singlets quenched by 30%

quenched by 80-90%

2 3 4 5 6 7

0.0

T

r

i

p

l

e

t

400 450 500 550 600

0

5000y

VWavelength /nm

Triplets are quenched very effectively in a driven device


Rapid loss of TTA during drivingRapid loss of TTA during driving

Model materials No triplet quencher

1

e

T100 T90

Device Lifetest from 5000Cdm-21.0

LuminanceTTA Ratio from fit

Model materials, No triplet quencher

0.1

d

L

u

m

i

n

a

n

c

e T80 T70 T60

0.8

0.9

e

/

T

T

A

R

a

t

i

o

TTA Ratio from fit

Ratio of EL from TTA drops

N

o

r

m

a

l

i

s

e

d

0.6

0.7

L

u

m

i

n

a

n

c

e

TTA drops rapidly compared to luminance

D l d fl l t d i l t f lif t t

0.0 2.0x10-7 4.0x10-7 6.0x10-7 8.0x10-7 1.0x10-60.01

Time /s

0.0 0.5 1.0 1.5 2.00.5

Time /h

Delayed fluorescence lost during early stages of lifetest


Effect of Triplet Quenching Additive

Effective reduction of triplet density

Effect of Triplet Quenching Additive

1 2 10-4

Effective reduction of triplet density when device is doped with triplet quenching additive.No change of emission spectrum

8.0x10-5

1.0x10-4

1.2x10 4 F8-PFB+Triplet Quencher F8-PFB

n

m

)

TQNo change of emission spectrum indicates that there is no singlet energy transfer20% drop in EQE with TQ consistent

2.0x10-5

4.0x10-5

6.0x10-5

d

T

/

T

(

7

8

0 TQ

20% drop in EQE with TQ consistent with loss of TTA contribution

2 3 4 5 6 7

0.0

V

T1 2 1 V

S1= 2.75eVS1=2.65eV

F8-PFB dPVBi

T1= 2.1eV T1= 2.0eVTriplet Quenching Additive: dPVBi (1% mol ratio)


( )

Lifetime with Triplet Quencher

1 0na

n

c

e

Lifetime with Triplet Quencher

0.8

1.0s

e

d

L

u

m

i

With Triplet Quencher

0.6

5 No

r

m

a

l

i

s

/

A Without

3

4

5

i

e

n

c

y

C

d

/ Without Triplet Quencher

0 1 2 3 4 5 62

E

f

f

i

c

Removing triplets reduces the rapid initial decay in lifetrace and givesRemoving triplets reduces the rapid initial decay in lifetrace and gives substantial improvements in stabilityLoss of TTA contribution to efficiency during early stages contributes significantly to the steep intitial slope of lifetrace


significantly to the steep intitial slope of lifetraceControl of triplet interactions is key for device efficiency and lifetime

p Oled Materials Device Operation

Documents

Transcript of p Oled Materials Device Operation