
Experimental characterization of photoemission from plasmonic nanogroove arrays
Christopher M. Pierce,1, 2, ∗Daniel B. Durham,3Fabrizio Riminucci,1Scott Dhuey,1
Ivan Bazarov,2Jared Maxson,2Andrew M. Minor,3and Daniele Filippetto1, †
1LBNL, 1 Cyclotron Road, Berkeley, California 94720, USA
2CLASSE, Cornell University, 161 Synchrotron Drive, Ithaca, New York 14853-8001, USA
3Department of Materials Science and Engineering, University of California, Berkeley, and National Center
for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720
(Dated: March 15, 2023)
Metal photocathodes are an important source of high-brightness electron beams, ubiquitous in the operation
of both large-scale accelerators and table-top microscopes. When the surface of a metal is nano-engineered
with patterns on the order of the optical wavelength, it can lead to the excitation and confinement of surface
plasmon polariton waves which drive nonlinear photoemission. In this work, we aim to evaluate gold plasmonic
nanogrooves as a concept for producing bright electron beams for accelerators via nonlinear photoemission.
We do this by first comparing their optical properties to numerical calculations from first principles to confirm
our ability to fabricate these nanoscale structures. Their nonlinear photoemission yield is found by measuring
emitted photocurrent as the intensity of their driving laser is varied. Finally, the mean transverse energy of
this electron source is found using the solenoid scan technique. Our data demonstrate the ability of these
cathodes to provide a tenfold enhancement in the efficiency of photoemission over flat metals driven with a
linear process. We find that these cathodes are robust and capable of reaching sustained average currents over
100 nA at optical intensities larger than 2 GWcm−2with no degradation of performance. The emittance of the
generated beam is found to be highly asymmetric, a fact we can explain with calculations involving the also
asymmetric roughness of the patterned surface. These results demonstrate the use of nano-engineered surfaces
as enhanced photocathodes, providing a robust, air-stable source of high average current electron beams with
great potential for industrial and scientific applications.
I. INTRODUCTION
High brightness electron sources for ultrafast applications
require prompt emission of high-charge electron beams and
direct injection into areas of extreme electromagnetic field
amplitudes. Photoemission from metal surfaces has been the
primary means of electron bunch generation, used by the large
majority of user facilities around the world [1–3], owing to
their fast response time and robustness. Despite their broad
use, metal cathodes have a few major disadvantages. First,
the typical quantum efficiency for a metal exhibits values in
the 10−5region which, for high charge pulse extraction, re-
quires laser pulse intensities close to the damage threshold
of the material. With time and continuous operation, this
has been shown to lead to partial ablation, increased surface
roughness, and reduced brightness [4]. High intensities may
also cause multi-photon absorption and photoemission, lead-
ing to the generation of unwanted halos, and an overall in-
crease of beam thermal emittance [5]. Furthermore, a typical
metal work function requires UV photons for linear photoe-
mission. The two-stage UV conversion from the initial in-
frared laser pulses has a substantial impact on the size and
complexity of the photocathode laser system. It may also im-
pact the quality of the final pulse, resulting in substantial loss
of energy, degradation of transverse pulse shape, and limited
control over longitudinal profile. Altogether, the low quan-
tum efficiency and the high work function effectively limit the
maximum average current that can be extracted by metal cath-
∗cmp285@cornell.edu
†dfilippetto@lbl.gov
odes and, therefore, the range of applications of the relevant
instrumentation.
High quantum efficiency semiconductor films provide a
possible path towards higher performance photocathodes. De-
pending on the choice of the material, the quantum efficiency
can be orders of magnitude larger for a work function in the
visible or infrared region [6]. Unfortunately, such cathodes
are chemically reactive, and the vacuum levels found in high
field photoinjectors often greatly complicate their use as high
brightness electron sources. Further, dark current may be-
come an issue in those same systems for materials with an
extremely low work function.
Nonlinear photoemission may offer another potential solu-
tion to avoid nonlinear wavelength conversion. Depending on
the material and laser parameters, it becomes more efficient to
extract electrons from the cathode directly via multi-photon
photoemission using infrared light, rather than perform wave-
length conversion to the UV [7]. However, as is the case for
linear photoemission, the small nonlinear yield of most flat
metallic surfaces demands laser fluence values close to the
material’s damage threshold (typically on the order of 0.1 to
1 Jcm−2[8]).
One path forward in improving the nonlinear yield of met-
als is by fabricating plasmonic structures by surface nanopat-
terning. Nanoscale grooves formed on a gold photocathode
have been shown to increase its nonlinear yield at 800 nm by
up to six orders of magnitude [9]. A similar concept using a
grid of nanoscale holes showed a dramatic increase in the non-
linear yield of gold and copper photocathodes [10, 11]. On the
other hand many questions remain open before such cathodes
could be effectively considered as a reliable source for ultra-
fast application: Can we produce nano-engineered cathodes
arXiv:2210.05056v2 [physics.acc-ph] 14 Mar 2023