Investigation of axial and transverse focal spot
sizes of Fresnel zone plates
TAO LIU,QIANG LIU,SHUMING YANG,* ZHUANGDE JIANG,TONG WANG, AND GUOFENG ZHANG
State Key Laboratory for Manufacturing System Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi 710049, China
*Corresponding author: shuming.yang@mail.xjtu.edu.cn
Received 6 February 2017; revised 23 March 2017; accepted 24 March 2017; posted 6 April 2017 (Doc. ID 286182); published 24 April 2017
An axial spot size formula is given for a standard binary Fresnel zone plate (FZP) illuminated with a linearly
polarized beam. The axial spot size, characterized by the full width at half-maximum (FWHM), can be accurately
expressed by the Jiang–Wilso n formula, viz., d
z
0.9λ
0
∕η − η
2
− NA
2
1∕2
, when the number of transparent
annuli is no less than 3. λ
0
is the illumination light wavelength, η is the refractive index, and NA is the equivalent
numerical aperture of a FZP. The transverse spot sizes are near linear with the radial width of the outmost annulus
when the cor responding numerical aperture is not very large. This study is carried out based on vectorial angular
spectrum theory and has been rigorously confirmed by the three-dimensional finite-difference time-domain
method. This finding provides the theoretical basis for applying FZPs in a variety of applications, such as nano-
lithography, nanoscopy, and high-density data storage.
© 2017 Optical Society of America
OCIS codes: (050.1380) Binary optics; (260.5430) Polarization; (050.1965) Diffractive lenses; (050.6624) Subwavelength structures.
https://doi.org/10.1364/AO.56.003725
1. INTRODUCTION
Fresnel zone plates (FZPs) suit a variety of applications, e.g.,
atomic optics, confocal imaging, x-ray nanoscopy, and synchro-
tron radiation experiments [1–4]. Recently, much attention has
been paid to superfocusing and nanoimaging of Fresnel zone
plates [5–8]. It is well known that the transverse (or lateral)
resolution of a low-numerical-aperture (NA) FZP has been sim-
ply expressed by d
t
1.22Δr
N
, according to the Rayleigh
criterion under simplified scalar diffraction theory [2]. Δr
N
is the radial width of the outmost annulus. In this situation,
the multifocus will appear at defined distances away from the
aperture plane of the FZP. However, for a relatively high-NA
FZP, the situation is quite different. There is only one main
focus [5–7] and all secondary foci have been suppressed to dis-
appear. This implies that for high-NA FZPs, there are no or
weak constructive interference positions between the FZP mask
and the main focus. To implement a complete study of a (high-
NA) FZP, the vectorial nature of the incident light source must
be considered, and suitable vectorial diffraction theories have to
be used instead of simplified scalar theory [7,9,10].
So far, there are no formulas that can be used to theoretically
describe the axial focal spot size of a standard binary FZP.
Additionally, there is also no formula that can characterize the
transverse spot size of the focus of a FZP with an arbitrary
numerical aperture. In this paper, the vectorial angular spectrum
(VAS) theory and the three-dimensional (3D) finite-difference
time-domain (FDTD) method are used to investigate the
formulas for axial and transverse focal spot sizes of Fresnel zone
plates. Thus, in the following section, the theoretical method for
analyzing the focusing performances of FZPs shall be intro-
duced. For a quick evaluation, the VAS theory can be used with
approximations, while the 3D FDTD method will then rigor-
ously validate the electr ic field distributions calculated by the
VAS theory.
2. METHOD
A. Fresnel Zone Plate
For a standard binary Fresnel zone plate, the radius coordinates
of each annulus are determined by [2]
r
n
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
nλf n
2
λ
2
∕4
q
;n 0; 1; 2; …;N; (1)
where f is the main foc al length and N is the total annulus
number. λ is the light wavelength in the medium where a FZP
is immersed. λ λ
0
∕η with λ
0
being the illumination light
wavelength and η the refractive index of the immersion
medium. The equivalent numerical aperture of a FZP is
defined by NA η sin α, similar to the case for a traditional
objective lens [11]. α is the maximum focusing semi-angle,
satisfying tan α r
N
∕f . The diameter of a FZP is thus
D 2r
N
2f · NA∕η
2
− NA
2
1∕2
.
It is known that when f ≫ λ, r
n
≈ nλf
1∕2
and the trans-
verse resolution, according to the Rayleigh criterion and scalar
diffraction theory, is approximated by [2]
Research Article
Vol. 56, No. 13 / May 1 2017 / Applied Optics 3725
1559-128X/17/133725-05 Journal © 2017 Optical Society of America