Molecular beam epitaxy of Cs-Sb thin films:

structure-oriented growth of high efficiency

photocathodes
Alice Galdi
Department of Industrial Engineering
Università degli Studi di Salerno (Italy)
CLASSE, Cornell University
 People

Maxson group Shen group

Jared Maxson Alice Galdi Kyle Shen Chris Parzyck Vivek Anil Hanjong Paik Brendan Faeth

Hines group
Cheng Hu Luca Moreschini
Chad Pennington Elena Echevarria

Arias group
Mellissa Hines Will DeBenedetti Jan Balajka

Darrel Schlom

Tomàs Arias J. Kevin Nangoi

https://cbb.cornell.edu/ https://www.paradim.org/
 Outline
• Motivation:
– Cs3Sb for brighter electron beams
– Surface disorder and emittance degradation

• Methods:
– RHEED monitored molecular beam epitaxy
• Results: epitaxy of Cs3Sb phase! (first time!)
– First band structure measurements of Cs3Sb thin films
• Results: new phase!
– Atomically smooth CsSb films: a visible light photoemitter with enhanced oxygen robustness
• Future work and conclusions
 Cs3Sb for low emittance, high current
Beam current: quantum efficiency, laser fluence, lifetime
2𝑚𝑒 𝑐2𝐼
𝐵𝑛 = 2
𝜎𝑥 𝑀𝑇𝐸 Mean transverse energy: Intrinsic Cs3Sb/Na2KSb: next generation
momentum spread + roughness + laser
spin polarized photocathode?
heating + …

λ=520 nm λ=690 nm, QE=0.7e-4

~0.3 mm mrad

~25 meV at 300 K

~8 meV at 90 K

B. Dunham et al. APL 102, 034105 (2013) L. Cultrera et al., Phys. Rev. STAB 18 (2015) 113401 V.S. Rusetsky et al, arXiv:2205.03693 [cond-mat.mtrl-sci]
 Drawbacks: reactive, disordered
λ=504 nm Why does this matter for applications?
Because both heterogeneity enhaced by
oxidation and roughness degrade MTE
(besides QE degradation in poor vacuum)

Δ𝑀𝑇𝐸 = 25 𝑚𝑒𝑉
@ 10 𝑀𝑉/𝑚
compare to measured:
42 𝑚𝑒𝑉@300K
σ=2.4 nm σ=1.75 nm 15𝑚𝑒𝑉@90𝐾

Measured surface and surface potential of a

Cs3Sb photocathode
G. S. Gevorkyan, PRAB 21, 093401 (2018)

A. Galdi, et al. The Journal of Chemical Physics 153,144705 (2020)

A Galdi, et al. Applied Physics Letters 118 (24), 244101
 Single crystalline photocathodes: epitaxy
Epitaxy is the growth of a crystal Epitaxial single-oriented films would allow:
layer with one or more well-defined • Roughness control
orientations with respect to an • Orientation control → surface potential control
underlying crystal seed layer • Measurements of intrinsic properties
(usually a single crystal substrate) (optical constants, band structure, intrinsic
MTE…)
Cs Sb
flux flux
1) Choose the method: MBE
2) Select suitable substrates
3) Identify suitable growth conditions

Epitaxial adsorption re-evaporation

layer

diffusion

Substrate nucleation
reaction-interdiffusion
 Previous results

Our work builds on many experimental results obtained

via in-operando characterization of the growth of alkali
antimonide thin films with different techniques.
 Cs3Sb structure and possible epitaxial relations
Jack and Wachtel (1957)
ത
𝐹𝑑3𝑚
Based on previous studies, we selected
3C-SiC as substrate
1x Sb +3x Cs
(but also tested: TiO2 , MgF2)
<100>

A Galdi, et al. Applied Physics Letters 118 (24), 244101

1x Sb +3x Cs

1x Sb +3x Cs
≈5%
1x Sb +3x Cs

Sb3- Cs+
1 X Cs3Sb (a=0.9165 nm)
4 X SiC (a=0.436 nm)

≈ ≈-1%
 RHEED assisted MBE @ PARADIM
PARADIM Thin Film User Facility
• Molecular Beam Epitaxy System Reflection High Energy Electron Diffraction
• In Operando high energy electron diffraction (RHEED)
• Pbase = 2x10-9 torr (RHEED)
• Sample Transfer System
• In Situ Quantum Efficiency Station (biased pickup coil)
• Pbase = 1x10-9 torr
• ARPES/XPS System Sample
• Scientia DA30 electron analyzer
• Fermi Helium Plasma discharge lamp 15kV e- gun
• Specs XR50 Al/Mg X-ray source
• Pbase = 7x10-11 torr Detector
MBE loadlock

Angle Resolved Photoemission Spectroscopy

(ARPES)
ARPES MBE
XPS w. STORAGE
LEED RHEED

Helium plasma
discharge lamp
Transfer chamber hv=21.22, 40.81 eV
(+X-ray tube
Mg/Al anode Hemispherical
QE measuring station
to UHV STM/XPS 1.254/1.487 keV) electron analyzer
vacuum suitcase
Epass = 2-100 eV
ARPES loadlock
 RHEED assisted MBE @ PARADIM
PARADIM Thin Film User Facility
• Molecular Beam Epitaxy System Reflection High Energy Electron Diffraction
• In Operando high energy electron diffraction (RHEED)
• Pbase = 2x10-9 torr (RHEED)
• Sample Transfer System
• In Situ Quantum Efficiency Station (biased pickup coil)
• Pbase = 1x10-9 torr
• ARPES/XPS System Sample
• Scientia DA30 electron analyzer
• Fermi Helium Plasma discharge lamp 15kV e- gun
• Specs XR50 Al/Mg X-ray source
• Pbase = 7x10-11 torr Detector

Angle Resolved Photoemission Spectroscopy

(ARPES)

Helium plasma
discharge lamp
hv=21.22, 40.81 eV
(+X-ray tube
Mg/Al anode Hemispherical
1.254/1.487 keV) electron analyzer
Epass = 2-100 eV
 The experiment
Typical photocathode growth: Our growth method:
Photocurrent monitored RHEED monitored
Quantum efficiency oriented Structure oriented

heater heater
Substrate Current Substrate
QMB
QMB

Light source
 Information provided by RHEED
Φ=0° • Real-time
Single crystal
• Sub-ML sensitivity
kin kout
• Qualitative probe of surface roughness and
crystallinity

Single crystal
High coherence
Φ=45°

Film
Reduced coherence
Roughened surface
Fluorescent screen

By rotating the sample Film

Θ around its surface normal, Polycrystalline
we intersect different sets domains/impurities
of reciprocal space rods
 Temperature study: codeposition
• Cs-Sb co-deposition on SiC substrates
Higher Growth Temperature
–Cs:Sb ratio = 6:1 (as measured by Quartz Crystal Microbalance)

λ=504 nm
Improved Crystal Order
OOP Film Orientation
Reduced Surface Roughness
Polycrystalline Reduced Quantum Efficiency
Fiber texture
Rough Flat

Lower Growth Temperature

Low crystallinity
Rough
Improved Quantum Efficiency
Polycrystalline Films
 Solid Phase (Molecular Beam) Epitaxy
• Low Deposition temperature to take advantage of
improved Cs sticking coefficient
• High temperature Cs anneal to improve crystallinity
T=85°C

T=40°C

Time
 It is epitaxial!

SiC [10] SiC [11]

Cs3Sb [10] Cs3Sb [11]

Φ=0° Φ=45°

e-
e-
 It really is Cs3Sb: XPS and ARPES

XPS: perfect agreement

ARPES: larger
bandwidth (Strain?
Structure distortion?)
C. W. Bates, et al. Thin Solid Films 69, 175 (1980)
High efficiency in the ultrathin limit

Grey line and dots

represent the data from a
codeposited Cs3Sb
sample grown by
photocurrent-monitored
codeposition.

Enhanced QE:
• At the photoemission
threshold
• At small thickness

C.T. Parzyck, A. Galdi et al.

Physical Review Letters 128, 114801
 Temperature study: codeposition
• Cs-Sb co-deposition on SiC substrates
Higher Growth Temperature
–Cs:Sb ratio = 6:1 (as measured by Quartz Crystal Microbalance)

λ=504 nm
Improved Crystal Order
OOP Film Orientation
Reduced Surface Roughness
Polycrystalline Reduced Quantum Efficiency
Fiber texture
Rough Flat

Lower Growth Temperature

Low crystallinity
Rough
Improved Quantum Efficiency
Polycrystalline Films
 Flat co-deposited samples

kin kout

Raw FFT

In vacuum
transfer to
0.6 nm Cs3Sb
FFT-filtered
STM
Lattice:
0.76 nm x 1.42 nm
 Flat co-deposited samples
CsSb parameters:
a = 7.34 Å
b = 7.57 Å
c =13.27 Å
P1 structure

Structure and composition data

𝐶𝑠: 𝑆𝑏 ≈ 1: 1 suggest that the compound we are
forming is closely related to CsSb
 Current work in PHOEBE @ Cornell U.
PHOEBE: PHOtocathode Epitaxy and Beam Experiments laboratory
Clean hood Vacuum suitcase Sample preparation

TE-meter Photocathode MBE

• Study of epitaxial Cs3Sb and

CsSb samples: spectral
response, lifetime, oxygen
resistance, mean transverse
energy

Chad Pennington Elena Echevarria

 Conclusions
• Epitaxy of Cs3Sb is achieved using molecular beam epitaxy via monitoring the sample
structure with RHEED
• State-of-the-art MBE machines and in-situ RHEED allow to explore various growth
regimes and efficient optimization of the samples beyond quantum efficiency
• High throughput:
– PARADIM experiments: 59 samples grown in 27 days (24h per day)
– PHOEBE: 75 Cs-Sb samples grown between 09/2021 and 08/2022
• Structure-oriented growth identifies an interesting phase: CsSb
– Atomically smooth
– Easily achievable via codeposition
– QE ~1% at 400 nm
– Robust against oxidation
 Information provided by RHEED
No texture
Epitaxial island growth Polycrystalline islands

kout kin kout

kin

Textured film
Transmission pattern:
Streaks turn into spots

Fiber texture

No rotation dependence if
the texture axis is out-of-
plane (uniaxial)
Rotation dependence if
the texture axis is in-
plane (biaxial)
 Epitaxial Relationship
Substrate Fiber Textured Film
Partially Ordered Film Epitaxial Film
Fully Ordered Only c-axis oriented