Extremely stable graphene electrodes doped with macromolecular acid

Although conventional p-type doping using small molecules on graphene decreases its sheet resistance (Rsh), it increases after exposure to ambient conditions, and this problem has been considered as the biggest impediment to practical application of graphene electrodes. Here, we report an extremely stable graphene electrode doped with macromolecular acid (perfluorinated polymeric sulfonic acid (PFSA)) as a p-type dopant. The PFSA doping on graphene provides not only ultra-high ambient stability for a very long time (> 64 days) but also high chemical/thermal stability, which have been unattainable by doping with conventional small-molecules. PFSA doping also greatly increases the surface potential (~0.8 eV) of graphene, and reduces its Rsh by ~56%, which is very important for practical applications. High-efficiency phosphorescent organic light-emitting diodes are fabricated with the PFSA-doped graphene anode (~98.5 cd A−1 without out-coupling structures). This work lays a solid platform for practical application of thermally-/chemically-/air-stable graphene electrodes in various optoelectronic devices.


Supplementary Note 1 | Surface morphology
Peak heights along the cross section of the PFSA-doped graphene were < 1 nm because deposition of thin polymeric layer can flatten uneven regions of the pristine graphene, such as wrinkles and grain boundaries; in contrast the peak heights on the pristine graphene were ~3 nm.

Supplementary Note 2 | Stability against acid and base chemicals
We performed stability test against acid to base chemicals. We dipped HNO3-and PFSA-doped  Fig. 6b). This result also proves the chemical stability of PFSA-doping on graphene.
Surface energies ( ) of graphenes were calculated by measuring contact angle ( ) of two solvents (DI water and diiodomethane), substituting the and constant of each solvent into Supplementary Eq.3, then solving the simultaneous equations for .

Supplementary Note 4 | Spatial uniformity investigation using c-AFM
We performed conductive atomic force microscopy (c-AFM) to investigate spatially resolve the electrical properties of graphene films according to doping methods ( Supplementary Fig.   9). In-plane conductance of HNO3-doped and PFSA-doped graphene were monitored after 5 days in ambient condition by measuring current between p-doped graphene and Sb-doped Si tip. p-Type doping with HNO3 or PFSA did not affect the surface topography of graphene.
PFSA-doped graphene film showed uniform current in surface mapping, the HNO3-doped graphene surface showed significant heteroheneity of in-plane conductance. This result confirms spatial uniformity and excellent doping stability of PFSA-doped graphene.

Supplementary Note 5 | Ambient stability investigation using Raman spectroscopy
After ambient exposure for 15 days, n of PFSA-doped graphene did not change, but n of HNO3doped graphene decreased drastically.

Supplementary Note 6 | Ambient stability investigation using Field-Effect Transistors
Ambient stability of PFSA doping was monitored by measuring the electrical characteristics of the FETs with PFSA-doped graphene ( Supplementary Fig. 11a,b). The upshift of the Dirac 23 point of PFSA-doped graphene did not decrease, and even slightly increased after the sample was held in ambient conditions for 21 days. Calculated average hole concentration (PFSA asdoped: 7.65 × 10 12 cm -2 , 7 days: 9.14 × 10 12 cm -2 , 21 days: 1.03 × 10 13 cm -2 ) also prove that PFSA doping of graphene is stable in ambient conditions (Supplementary Fig. 11c).  Table 4); this result shows that chemical doping of graphene using macromolecules would lead to high binding energy and impart excellent doping stability to graphene.

Supplementary Note 8 | X-ray photoelectron spectroscopy
We performed X-ray photoelectron spectroscopy (XPS) to investigate the change of chemical composition by PFSA doping. Pristine and PFSA-doped graphene showed an O1s peak that could be caused by oxygen in the glass substrate, or in PMMA left after the wet-transfer process.

Supplementary Note 10 | Control measurements using PEDOT:PSS
PEDOT:PSS was spin-cast on pristine graphene layer without any treatment to make graphene surface hydrophilic. Film uniformity of spin-cast PEDOT:PSS on hydrophobic graphene surface was poor, and left may uncoated regions ( Supplementary Fig. 15a, b). Raman spectra  Fig. 15d); the upshifted G band position then downshifted from 1607 cm -1 to 1597 cm -1 ; this change indicates that the p-type doping effect had weakened. The change occurs because PSS is hygroscopic, so it takes up moisture from ambient air and thereby degrades the p-type doping effect of PEDOT:PSS on graphene.

Supplementary Note 11 | Hole-injection efficiency calculation
We calculated hole-injection efficiency (η) by using theoretically-calculated space-charge limited current (SCLC) as followed by Supplementary Eq. 4.

Supplementary Note 12 | Polymeric HILs deposition on PFSA-doped graphene
To deposit polymeric hole-injection layers (HILs) 3 , we used perfluorinated ionomer (PFI)blended PEDOT:PSS solution (we call it GraHIL), which contains isopropyl alcohol (IPA) rather than as-purchased PEDOT:PSS solution, because it cannot be uniformly coated on hydrophobic graphene surface ( Supplementary Fig. 17a). The OLEDs were fabricated with a GraHIL, which has high surface WF ~5.95 eV. [4][5][6] Diluting with IPA significantly reduced the contact angle of polymeric solution on graphene surface, because the lower polarity of IPA than that of H2O improved the wettability. However, diluting with IPA was still insufficient to form uniform polymeric HILs on graphene surface. Blending with PFI as well as addition of IPA to PEDOT:PSS solution yielded a uniformly-cast PEDOT:PSS film on hydrophobic graphene surface. To make the graphene surface hydrophilic, we also performed mild UV-O3 treatment before spin-casting the PFI-blended PEDOT:PSS solution. These attempts resulted in a uniform polymeric HIL film on graphene electrode ( Supplementary Fig. 17b). PFSA-doped graphene has even lower surface energy (21.69 mJ m -2 ) than pristine graphene (55.73 mJ m -2 ) and therefore has poor film-formability ( Supplementary Fig. 17c). However, improved wettability from IPA dilution, and PFI blending works in PFSA-doped graphene, so uniform polymeric HILs formed on PFSA-doped graphene surface ( Supplementary Fig. 17d).

Supplementary Note 13 | Device characteristics with PFSA-doped graphene anodes
The OLED with the PFSA-doped graphene anode exhibited a higher current density than did the OLED with pristine 4LG anode; this result was also caused by improved hole injection from graphene anode due to the increased surface WF of PFSA doped graphene ( Supplementary Fig.   18a). The device with PFSA-doped 4LG also showed higher power efficiency (PE) (~95.6 lm W -1 without an outcoupling structure) than did the device with the pristine 4LG (~77.6 lm W -