Engineering Ultra-Low Work Function of Graphene
Hongyuan Yuan,*
,†
Shuai Chang,
‡
Igor Bargatin,
#
Ning C. Wang,
‡
Daniel C. Riley,
†
Haotian Wang,
∥
Jared W. Schwede,
†
J. Provine,
‡
Eric Pop,
‡
Zhi-Xun Shen,
†,∥
Piero A. Pianetta,
‡,⊥
Nicholas A. Melosh,
§
and Roger T. Howe
‡
†
Department of Physics,
‡
Department of Electrical Engineering,
§
Department of Material Science and Engineering, and
∥
Department
of Applied Physics, Stanford University, Stanford, California 94305, United States
⊥
Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, 2575 Sand Hill Road, MS31, Menlo Park,
California 94205, United States
#
Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania, Philadelphia, Pennsylvania 19104,
United States
* S Supporting Information
ABSTRACT: Low work function materials are critical for energy conversion and electron emission applications. Here, we
demonstrate for the first time that an ultralow work function graphene is achieved by combining electrostatic gating with a Cs/O
surface coating. A simple device is built from large-area monolayer graphene grown by chemical vapor deposition, transferred
onto 20 nm HfO
2
on Si, enabling high electric fields capacitive charge accumulation in the graphene. We first observed over 0.7
eV work function change due to electrostatic gating as measured by scanning Kelvin probe force microscopy and confirmed by
conductivity measurements. The deposition of Cs/O further reduced the work function, as measured by photoemission in an
ultrahigh vacuum environment, which reaches nearly 1 eV, the lowest reported to date for a conductive, nondiamond material.
KEYWORDS: Graphene, work function, electrostatic gating, transistor, photoemission, scanning Kelvin probe force microscopy
T
he work function (Φ) of a material is the energy
difference between its vacuum level and Fermi level (E
F
).
It is not a fundamental constant but can be tuned through
doping or surface engineering. Materials with very low work
function can significantly improve many electronic device
technologies including organic electronics
1−4
and electron
emission devices.
5−11
Similarly, recently proposed solar energy
conversion technologies are predicted to have very high
efficiencies if sufficiently low work function anodes can be
produced.
12
To lower the work function of a material, one typical
approach is to lower the vacuum level by surface engineering,
particularly by depositing a very thin layer (about one
monolayer) of alkali metal, such as Cs, Li, Sr, or Ba,
13−16
that are sometimes combined with a proper amount of oxygen.
Among these approaches, Cs/O coated materials typically have
the lowest work function between 1.1 and 1.4 eV.
17−19
The
underlying mechanism for the work function reduction through
the application of a thin layer of Cs/O has been extensively
studied in the 1960s and 1970s, primarily driven by the
development of negative electron affinity (NEA) photo-
cathodes.
17,20−22
Another approach to lowering the work function is to raise
the material’s Fermi level. In contrast to conventional three-
dimensional materials, whose Fermi-level is normally “pinned”
at the surface due to surface defects and traps,
23
the Fermi level
of graphene (a two-dimensional material) can be effectively
controlled by doping due to a lack of dangling bonds and
surface states. By voltage biasing graphene relative to a gate,
compensating charges build up in the graphene. This excess
population of carriers shifts the Fermi level relative to its
equilibrium value, thereby directly changing the graphene work
Received: May 15, 2015
Revised: September 22, 2015
Published: September 24, 2015
Letter
pubs.acs.org/NanoLett
© 2015 American Chemical Society 6475 DOI: 10.1021/acs.nanolett.5b01916
Nano Lett. 2015, 15, 6475−6480