Nature of Non-emissive Black Spots in Polymer LEDs Ji-Seon Kim, Peter K. H. Ho, Craig E. Murphy, Nicholas Baynes, and Richard H. Friend Reviewed by Joung-Mo.

Nature of Non-emissive Black Spots in Polymer

LEDs

Ji-Seon Kim, Peter K. H. Ho, Craig E. Murphy, Nicholas Baynes, and Richard H. Friend

Reviewed by Joung-Mo Kang for 6.977, Spring 2002

S. H. Kim et al Synthetic Metals 111-112 (2000) 254

McElvain et al. J. Appl. Phys., Vol. 80, No. 10, 15 Nov 1996 6004

The Phenomenon ObservedThe Great Organics Plague

ExperimentTest PLED materials

•poly(4-styrenesulfonate)-doped poly(3,4-ethylenedioxythiophene) = PEDOT:PSS

•poly(2,7-(9,9-di-n-octylfluorene-alt-benzothiadiazole)) = F8BT

•poly(2,7-(9,9-di-n-octylfluorene)-alt-(1,4-phenylene-((4-sec-butylphenyl)imino)-1,4-phenylene)) = TFB

50:50 F8BT:TFB – 80nm

ExperimentDevice Structure

Al – 400nm

Ca – 5nm

7% PEDOT in PSSH – 50nm

ITO – substrate

•Eight 16mm² LEDs fabricated on patterned ITO substrate•Encapsulated with a cover glass and epoxy resin•Emit yellow-green•Low drive voltage, high current density (>100mA/cm², 3V)•High power efficiency (>20lm/W)•Lifetime exceeds 5000h at 100 cd/m²

ExperimentDevice Characteristics and Experimental Conditions

Devices were driven in ambient atmosphere at room temp for 120h with J = 100 mA/cm² and initial brightness L = ~104 cd/m² Top left figure is an optical picture taken in reflected light. Two ~2 mwide pinholes + disks are visible in each of the glass and ITO areas of substrate. Bottom shows same device turned on. The term “black spots” describes this dark patch in the yellow-green EL emission.

     Raleigh Scatter      Raman Scatter

Raleigh wavelength same as incident, Raman wavelength is different

•For a given monochromatic incident beam, there will be many frequencies of Raman-scattered light•The difference in energy of the incident and scattered light is the Raman shift, and is associated with some coupled molecular vibrational mode•A Raman spectrum depends on the molecule and its environment, however:•The Raman shifts are independent of the frequency of the exciting light

AnalysisIntroduction to Raman Scattering (extremely abridged)

• Non-destructive

• Can detect beyond glass/ITO layers at appropriate frequencies

• Can tune excitation frequency for greater response to molecules or structures of interest

• 10x greater spatial resolution than FTIR (~0.5 m at = 633 nm vs ~5 m at = 4-10 m)

• Shifts can indicate conjugation length changes

AnalysisAdvantages of Raman Spectroscopy

DataRaman Spectra

DataInterpretation

1. Away from defect, spectra indicate a combination of polymer blend and doped PEDOT as expected

2. Within defect, PEDOT becomes “dedoped” (reduced)

3. Emissive polymers appear not to migrate or to suffer damage

4. Metal oxide formation within disc, outside of pinhole

5. Dedoping method is passive: defects formed over glass where no current was injected

DiscussionProposed Mechanism

DiscussionSo What Does It All Mean?

• Non-emissive discs of reduced PEDOT and metal oxide form around pinhole defects in the cathode

• Each half of this redox reaction produces a non-conducting material, cutting off local current density

• Thus black spots reduce device active area and total luminescence output, but not EL efficiency

• The drop in efficiency that is observed is due to other mechanisms such as interfacial degredation

ComparisonWhere This Paper Fits Into the Current Canon

• It is widely agreed that pinhole defects source a disc-shaped black spot in many organic devices, and that these defects are only formed during manufacture

• Many papers found oxidation of the metal at organic interfaces causing loss of EL, or that spots are caused by a lack of carrier injection rather than quenching

• One other paper agrees that loss of luminescence is intrinsic to device and independent of black spots

• Several theories were specifically refuted as well, such as the dependence of black spot formation on carrier injection or conjugation length changes

ComparisonSome Other (Possible) Degradation Defects

• Gas evolution, metal bubbles

• Bright-ringed, non-circular black spots

• “Self-healing” point defects

• Crystallization of organics

CriticismInquiring Minds Want to Know

• What happens without a low work function, positively charged dopant like PEDOT?

• What about the many findings of water and oxygen oxidizing metal interfaces on their own?

• Time-varying data?