Optimization and Characterization of Nanostructured Surfaces for Photon-Enhanced Thermionic Emission and Photoemission Cathodes D.C. Riley1,2*, V.K. Narasimhan1,2, J. Jean3, I. Bargatin3, J.W. Schwede1,2 R.T. Howe3, Z.-X. Shen1,2, N.A. Melosh1,2
Photon-Enhanced Thermionic Emission
We recently proposed a new method of solar energy harvesting which for the first time combines quantum and thermal mechanisms into a single process. The new method, called photon-enhanced thermionic emission (PETE), can potentially overcome the losses inherent to PV while bypassing the challenges faced by traditional thermionics. Solar photons excite carriers into the conduction band of a p-type semiconductor. These photo-excited carriers rapidly thermalize within the conduction band, and if they encounter the surface with sufficient thermal energy, they can escape into vacuum.
Photoexcitation boosts the thermionic emission current over the equilibrium level by increasing the population of electrons in the conduction band. The emission current can be described by the R i c h a r d s o n - D u s h m a n equation where the work function is defined relative to the conduction-band quasi-Fermi level:
2*
eff kTPETE
kTe x
J A T eq n v e
φ
χ
−
−
=
=
Nanostructuring to Enhance Photocathodes Revisiting Spicer’s 3-Step Model
Photoemission Simulation Suite
Acknowledgements and Author Affiliations
*[email protected], 1Geballe Laboratory for Advanced Materials 2Stanford Institute for Materials and Energy Sciences 3Center for Interfacial Engineering for Mechanical Systems at Stanford
Theoretical Device Performance
PETE vs. Ideal PV PETE vs. PETE+Thermal
Theoretical PETE conversion efficiency as a function of Eg and concentration of the direct+circumsolar AM1.5 solar spectrum has been calculated. Each Photon with energy greater than the bandgap is absorbed and creates one electron/hole pair, while sub-bandgap photons are absorbed as heat. Black body radiation losses and transfer, as well as radiative and Auger recombination losses are included. The cathode temperature and electron affinity are chosen to maximize overall efficiency. A PETE device combined with solar thermal could have efficiencies >50%!
J. W. Schwede et al., Nature Materials, 9, 762–767 (2010).
Detectors Electron Guns Energy Converters
Photocathodes are surfaces that emit electrons when illuminated with photons due to the photoelectric effect. They are usually a metal or a semiconductor coated with an alkali metal to increase the efficiency by lowering the work function. Traditionally photocathodes have been used in high-sensitivity detectors, like night-vision goggles, as well as in electron guns for particle accelerators, with potential extensions into electron beam lithography and scanning electron microscopy. Our group has also been investigating the use of photocathodes for solar energy conversion devices. Such a device would require a very high quantum efficiency (fraction of absorbed photons converted to collected electrons) approaching 100% for reasonable power conversion efficiency, similar to photovoltaics.
Increased Absorptivity There is significant room for improvement of photoemission based devices, because they traditionally use a planar geometry which is not ideal for absorbing light. Nanostructures have been used in photovoltaic applications to increase absorptivity by introducing a smooth gradient in the effective index of refraction at t h e a b s o r p t i v e s u r f a c e . Absorptivity measurements show improvement over thin films for all angles.
J. Zhu et. al., Nanoletters, 2008
Confinement of Carriers Generating and confining carriers near the emissive surface is important for photocathodes and particularly important for a PETE based device, because a carrier may encounter the surface many times before it emits. Nanostructuring may enable us to decouple the emission and absorption lengths, thus avoiding the competition between absorption and emission normally associated with a planar geometry.
Absorption Transport Transport Emission Escape
“Probability of escaping after being emitted”
The photoemission process is best understood through Spicer’s 3-step model which consists of absorption, transport, and emission. However Spicer’s model only considers a planar geometry, therefore emitted carriers are assumed to escape the surface. If a nanostructured geometry is considered then emitted carriers are no longer guaranteed to escape the surface structure. A 4th step in the photoemission process must be considered which is the probability of escaping after being emitted. We have created a simulation suite to understand the full photoemission process in nanostructures.
Reflectivity Diffusivity Angular Distribution
Sim
ulat
ion
Tool
s E
xper
imen
t
Bias Dependence
Grating 350 nm Width 700 nm Pitch
Grating 650 nm Width 700 nm Pitch
Planar Surface
1 x 104
0
5 x 103
Phot
ons/
(cm
3 s)
1400 nm
5x105 V/m Bias
80%
No Bias
35% Absorp'on (AM 1.5D+C) 0.64 0.77 0.88 Transport and Emission 0.83 0.85 0.86 Escape 1.00 0.99 0.61 Yield 0.53 0.65 0.46
Comparison of Negative Electron Affinity (NEA) GaAs Tapered Grating Simulation Results
Planar Tapered Grating
350 nm Base Tapered Grating
650 nm Base
700 nm pitch
Photocathode Applications
Absorp'on (AM 1.5D+C) 0.64 0.78 0.998 Transport and Emission 0.83 0.76 0.86 Escape 1.00 0.96 0.18 Yield 0.53 0.57 0.15
400x1600 nm 200 base x 320 height nm
Comparison of NEA GaAs Nanocone Simulation Results
Absorption Transport Emission Escape
Light absorption is calculated using the Rigorous Couple-Wave Analysis method for 3D structures and the Finite-Difference Frequency Domain method for 2D Structures.
The transport step is calculated using a Monte Carlo simulation which takes into account the full band structure of the material. Carrier excitation information is provided by the absorption simulation step.
The emission step is strongly coupled with the transport step, therefore they are calculated together in the Monte Carlo simulation
The escape step is calculated using SIMION. Position and trajectory information is input from the Monte Carlo. The electric fields are calculated as a function of geometry and bias. Carriers are allowed to fly ballistically. The results show that there is a tradeoff between absorption and escape probability.
The structure dimensions must be chosen to increase the absorption without harming the escape probability too much. Note that in GaAs most electrons emit nearly perpendicular to the surface due to the low crystal mass. The aspect ratio and spacing of the structures can be chosen to take advantage of this effect.
Experimental Results Nanostructure Absorptivity
GaAs, p-doped, 5x1018 cm-3 Measurements were performed on p-doped 5x1018 cm-3 GaAs using normally incident unpolarized light. The nanocones and the taepered grating structure, shown in the SEM images to the left, exhibited significantly improved absorptivity across all measured wavelengths. The nanocones and the tapered structure exhibited high quantum efficiency, however they didn’t exceed the planar geometry. One possible route for improvement is to increase the quality of the surface by optimizing the etch chemistry. Another route for improvement is optimizing the Cs-O activation procedure. The activation procedure for planar samples is well understood from years of research, whereas nanostructures activate quite differently and are not well understood.
Nanostructure Photoemission GaAs, p-doped, 5x1018 cm-3, Cs-O coating
Planar