U.S. patent application number 14/862328 was filed with the patent office on 2017-03-23 for diffractive optical element and method for the design of a diffractive optical element.
This patent application is currently assigned to STMicroelectronics (Research & Development) Limite. The applicant listed for this patent is STMicroelectronics (Research & Development) Limite. Invention is credited to James Peter Drummond Downing.
Application Number | 20170082862 14/862328 |
Document ID | / |
Family ID | 57334884 |
Filed Date | 2017-03-23 |
United States Patent
Application |
20170082862 |
Kind Code |
A1 |
Downing; James Peter
Drummond |
March 23, 2017 |
DIFFRACTIVE OPTICAL ELEMENT AND METHOD FOR THE DESIGN OF A
DIFFRACTIVE OPTICAL ELEMENT
Abstract
A diffractive optical element (DOE) is designed to implement
both a collimation function with respect to an input divergent beam
and a beam shaping function with respect to an output divergent
beam. The phase designs of the collimation function and the beam
shaping function are independently produced in the phase domain.
These phase designs are then combined using a phase angle addition
of the individual functions and wrapped between 0 and 2.pi.
radians. The diffractive surface of the DOE is then defined from
the wrapped phase angle addition of the individual functions.
Inventors: |
Downing; James Peter Drummond;
(Doune, GB) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
STMicroelectronics (Research & Development) Limite |
Marlow |
|
GB |
|
|
Assignee: |
STMicroelectronics (Research &
Development) Limite
Marlow
GB
|
Family ID: |
57334884 |
Appl. No.: |
14/862328 |
Filed: |
September 23, 2015 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G02B 27/0012 20130101;
G02B 27/0927 20130101; G02B 27/30 20130101; G02B 27/0944
20130101 |
International
Class: |
G02B 27/09 20060101
G02B027/09; G02B 27/00 20060101 G02B027/00; G02B 27/30 20060101
G02B027/30 |
Claims
1. An optical system, comprising: a collimating and beam shaping
diffractive optical element configured to modify an input divergent
beam and produce an output divergent beam; wherein the collimating
and beam shaping diffractive optical element comprises a
diffractive surface having a design which both collimates the input
divergent beam and shapes the collimated input divergent beam into
the output divergent beam to form an output field with a desired
output intensity distribution.
2. The optical system of claim 1, wherein said diffractive surface
is formed by a surface relief pattern designed to implement both a
collimation function and a beam shaping function.
3. The optical system of claim 2, wherein the surface relief
pattern is defined by a phase combination of a phase profile for
collimation of the input divergent beam and a phase profile for
shaping the output divergent beam.
4. The optical system of claim 3, wherein the surface relief
pattern is further defined by a conversion of the phase combination
into a physical surface relief profile.
5. The optical system of claim 3, wherein the phase profiles for
collimation and shaping are quantized phase profiles.
6. The optical system of claim 1, wherein said diffractive surface
is formed by a gradient-index (GRIN) material designed to implement
both a collimation function and a beam shaping function.
7. The optical system of claim 6, wherein the GRIN material is
defined by a phase combination of a phase profile for collimation
of the first divergent beam and a phase profile for shaping the
second divergent beam.
8. The optical system of claim 7, wherein the GRIN material is
further defined by a conversion of the phase combination into a
GRIN structure.
9. The optical system of claim 6, wherein the phase profiles for
collimation and shaping a quantized phase profiles.
10. A method, comprising: defining a first phase profile for
collimation of an input divergent beam; defining a second phase
profile for shaping an output divergent beam; adding the first and
second phase profiles to form a combined phase profile; and forming
a diffractive surface of a diffractive optical element from the
combined phase profile so that the diffractive surface of the
diffractive optical element is configured to both collimate the
input divergent beam and shape collimated input divergent beam into
the output divergent beam to form an output field with a desired
output intensity distribution.
11. The method of claim 10, wherein said diffractive surface is
formed by a surface relief pattern defined from the combined phase
profile to implement both a collimation function and a beam shaping
function.
12. The method of claim 11, wherein adding comprises summing a
first phase profile for the collimation function and a second phase
profile for the beam shaping function.
13. The method of claim 12, wherein the first and second phase
profiles are quantized phase profiles.
14. The method of claim 10, wherein said diffractive surface is
formed by a gradient-index (GRIN) material defined from the
combined phase profile to implement both a collimation function and
a beam shaping function.
15. The method of claim 14, wherein adding comprises summing a
first phase profile for the collimation function and a second phase
profile for the beam shaping function.
16. The method of claim 15, wherein the first and second phase
profiles are quantized phase profiles.
17. A method, comprising: independent design of a first phase
profile for a collimation function and a second phase profile for a
beam shaping function; combination of the independently designed
first and second phase profiles using a phase angle addition;
wrapping of the phase angle addition between 0 and 2.pi. radians;
and production of a physical optic using a diffractive surface
defined by the wrapped phase angle addition.
18. The method of claim 17, wherein production comprises formation
of a surface relief pattern designed to implement both the
collimation function and the beam shaping function.
19. The method of claim 17, wherein production comprises formation
of a gradient-index (GRIN) material designed to implement both the
collimation function and the beam shaping function.
20. The method of claims 17, wherein the independent design
comprises independently designing the collimating function and the
beam shaping function each in a phase space; wherein combination
comprises summing in the phase space the collimating function and
the beam shaping function; wherein production comprises forming a
diffractive surface from the summation in phase space that will
perform both the collimating and beam shaping functions.
21. The method of claim 20, wherein the diffractive surface
comprises a surface relief pattern.
22. The method of claim 20, wherein the diffractive surface
comprises a gradient-index (GRIN) material.
Description
TECHNICAL FIELD
[0001] The present invention relates to beam shaping devices and,
in particular, to a diffractive optical element (DOE) and a method
for designing a DOE to reduce or eliminate the risk of high
intensity zeroth order leakage.
BACKGROUND
[0002] Reference is made to FIG. 1 showing a conventional geometry
for an optical system 10. The system 10 includes a light source 12
generating a collimated beam 14 of light with a planar wavefront
that is propagated toward a beam shaping diffractive optical
element 16. The diffractive optical element (DOE) 16 is a shaping
(for example, circularly homogenizing) optical element designed to
generate a desired non-collimated output beam 18 forming an output
field 20 with a desired output intensity distribution. Due to small
errors in the diffractive surface of the diffractive optical
element 16 (e.g., surface defects introduced during manufacture),
there is possibility for a high intensity zeroth order leakage 22
of the collimated beam 14 of light through the diffractive optical
element 16 into the output field 20. In this context, reference to
"high intensity" means dangerous in that the leakage has a beam
divergence and power that is dangerous (see, for example, British
standard BS EN 60825-1:2014 or international standard IEC
60825-1:2014, both incorporated by reference).
[0003] Reference is made to FIG. 2 showing a conventional geometry
for an optical system 10'. The system 10' includes a light source
12' generating a divergent beam 26 of light that propagates toward
a collimating lens 28. The collimating lens 28 functions to
collimate the divergent beam 26 and output a collimated beam 14 of
light with a planar wavefront that propagates toward a beam shaping
diffractive optical element 16. The diffractive optical element 16
is a shaping (for example, circularly homogenizing) optical element
designed to generate a desired non-collimated output beam 18
forming an output field 20 with a desired output intensity
distribution. Due to small errors in the diffractive surface of the
diffractive optical element 16 (e.g., surface defects introduced
during manufacture), there is possibility for a high intensity
zeroth order leakage 22 of collimated beam 14 of light through the
diffractive optical element 16 into the output field 20.
[0004] The high intensity zeroth order leakage 22 in FIGS. 1 and 2
is ostensibly an unmodified portion of the collimated beam 14
present in the output field 20. The effect of this is illustrated,
for example, in FIG. 3 which shows an image (with inverted
contrast) of a square homogenized output field 20 that is corrupted
by high intensity zeroth order leakage 22. If the light source 12
or 12' is, for example, a laser light source, this zeroth order
leakage 22 is a safety risk that will have a direct effect on the
permitted operating power of the light source 12 or 12'. In
response the existence of or chance for such leakage, the power
level of the light source 12 or 12' must be reduced to ensure safe
operation, and no benefit is gained from using the diffractive
optics.
[0005] Evaluation of the beam shaping diffractive optical element
16: There are many sources of error that can degrade the form
factor of the diffractive optic with respect to the initial design
form. If magnitudes of these errors are set to limits expected in
manufacture, the error in the height of the module (i.e., the depth
error) is recognized as the key contributor to degradation of optic
performance. FIGS. 3A-3C illustrate simulated results of the
element 16 with a nominal depth error (i.e., less than or equal to
5%) in FIG. 3A, a 10% depth error in FIG. 3B and a 15% depth error
in FIG. 3C. It will be noted that with more significant depth
errors homogenization is affected the zeroth order leakage
characteristic (reference 22) becomes more dominant.
[0006] Diffractive optics allow the optical system designer to
manipulate a wavefront in ways that cannot be achieved with
refractive optics. This allows a greater flexibility of the
functionality of the optic and enables applications that are not
supported by refractive optics. There is a need, however, for an
improved diffractive optical element that is not susceptible to
surface errors which would permit passage of high intensity zeroth
order leakage. As a result, operation of the optical system at
higher power levels would be possible.
SUMMARY
[0007] In an embodiment, an optical system comprises: a collimating
and beam shaping diffractive optical element configured to modify
an input divergent beam and produce an output divergent beam;
wherein the collimating and beam shaping diffractive optical
element comprises a diffractive surface having a design which both
collimates the input divergent beam and shapes the collimated input
divergent beam into the output divergent beam to form an output
field with a desired output intensity distribution.
[0008] In an embodiment, a method comprises: defining a first phase
profile for collimation of an input divergent beam; defining a
second phase profile for shaping an output divergent beam; adding
the first and second phase profiles to form a combined phase
profile; and forming a diffractive surface of a diffractive optical
element from the combined phase profile so that the diffractive
surface of the diffractive optical element is configured to both
collimate the input divergent beam and shape collimated input
divergent beam into the output divergent beam to form an output
field with a desired output intensity distribution.
[0009] In an embodiment, a method comprises: independent design of
a first phase profile for a collimation function and a second phase
profile for a beam shaping function; combination of the
independently designed first and second phase profiles using a
phase angle addition; wrapping of the phase angle addition between
0 and 2.pi. radians; and production of a physical optic using a
diffractive surface defined by the wrapped phase angle
addition.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Other advantages and characteristics of the invention will
become apparent on studying the detailed description of
embodiments, which are taken by way of non-limiting examples and
are illustrated by the appended drawings, in which:
[0011] FIGS. 1 and 2 show conventional geometries for optical
systems;
[0012] FIG. 3 is an image illustrating the presence of high
intensity zeroth order leakage in the output field generated by the
systems of FIGS. 1 and 2 due to the presence of an error on the
surface of the diffractive optical element;
[0013] FIGS. 3A-3C illustrate simulated results of a conventional
diffractive lens profile design with a variety of depth errors
[0014] FIG. 4 shows a geometry for an embodiment of an optical
system;
[0015] FIG. 5 illustrates the relationship between wavefront phase
and radial distance for a point source illumination at a finite
distance;
[0016] FIG. 6 illustrates a quantized example of a diffractive lens
profile design;
[0017] FIGS. 7A-7B illustrate examples for beam shaping optic
design at different quantizations;
[0018] FIG. 7C shows the quantized beam shaping optic of FIG. 7B
tessellated to fill and area equivalent to the area of the lens
profile of FIG. 6;
[0019] FIG. 8 illustrates the result of summing the phase profiles
of a diffractive lens from FIG. 6 and a beam shaping optic such as
that shown in FIG. 7C;
[0020] FIG. 9 illustrates the result of wrapping the summed phase
profiles between 0 and 2.pi. radians; and
[0021] FIGS. 10A-10C illustrate simulated results of the
diffractive lens profile design with a variety of depth errors.
DETAILED DESCRIPTION
[0022] Reference is now made to FIG. 4 showing a geometry for an
optical system 100. The system 100 includes a light source 112
generating a divergent beam 126 of light that propagates toward a
collimating and beam shaping diffractive optical element 116. The
diffractive optical element (DOE) 116 performs two functions: a)
the DOE collimates the input divergent beam 126 of light; and b)
the DOE shapes the light to generate a desired divergent
(non-collimated) output beam 118 forming an output field 120 with a
desired output intensity distribution (for example, homogenized
with a certain shape).
[0023] The advantage of the DOE 116 is that any surface error that
would inhibit the correct operation of the beam shaping optic will
also inhibit the correct operation of the collimating function. As
a result, the DOE 116 impedes output of a high intensity zeroth
order leakage (for example, collimated) beam in the presence of
surface errors and permits higher power operation of the light
source 112 in comparison to the systems of FIGS. 1 and 2 if exposed
to an equivalent surface error or defect.
[0024] The design of the DOE 116 utilizes a process that includes:
a) independent design of a phase-only diffractive lens profile and
a phase-only beam shaper optic; b) combination of the two
independent designs using a phase angle addition of their
individual functions; and c) production of the physical optic, for
example the diffractive surface, that will perform the function of
the phase-combined designs. The combined diffractive lens and beam
shaping designs are preferable implemented in a single diffractive
optic surface of the DOE 116. The independent or decoupled
optimization of the collimation function and the beam shaping
function during phase-only design simplifies the optimization
process and improves the efficiency and uniformity of the output
field 120. The collimating and beam shaping functions are first
independently defined in the phase space, with those functions then
summed in the phase space (for example, using phase angle addition)
and converted to the physical to yield a single surface profile for
the optic that will perform both the collimating and beam shaping
functions. The advantage of this implementation derives from the
combined effect a manufacturing error will have on the performance
of the collimating and beam shaping functions. Any surface profile
error in the optic that deviates from the nominal design will
impede the ability of the DOE 116 optic to perform both functions.
Therefore, with an error, the ability of the optic to form an
undesirable high intensity collimated beam (such as with high
intensity zeroth order leakage) is degraded in the same way as the
ability of the optic to form the desired beam shape. As a result,
there is a significant reduction in the sensitivity of the optic to
manufacturing error (especially in comparison to the conventional
approaches as shown in FIGS. 1 and 2 where the beam is discretely
collimated and the inclusion of a surface defect may permit passage
of a component of the collimated beam as high intensity zeroth
order leakage).
[0025] In a preferred implementation, the design of the diffractive
lens profile must have a focal length equal to a desired separation
distance d between the DOE 116 and the light source 114.
Furthermore, the design of the beam shaper function must assume
that the input light is collimated.
[0026] Diffractive lens profile design: The design of a diffractive
lens is well known to those skilled in the art. Reference is made
to the textbook "Computer design of diffractive optics," Soifer,
Elsevier, 2012 (incorporated by reference). For the DOE 116, a lens
is designed based on a necessity to correct the curvature of a
wavefront from a point source (light source 114) at a finite
distance d from the optic. The optic is defined such that its focal
length is equivalent to the separation distance d of the point
source from the optic. See, for example, FIG. 5.
[0027] The lens profile 0(x,y)=exp[i*.phi.(x,y)] is designed to add
an equivalent opposite phase delay to the wavefront profile
W(x,y)=exp[i*.phi.(x,y)] so that it matches the target phase
profile P(x,y)=exp[i*.xi.(x,y)]:
P(x,y)=W(x,y)+0(x,y)
[0028] For a collimating lens, the target phase profile P(x,y)=1
for all x and y:
0(x,y)=P(x,y)-W(x,y)=1-W(x,y)
[0029] Which is equivalent to:
.phi.(x,y)=.xi.(x,y)-.phi.(x,y)=-.phi.(x,y)
[0030] Therefore, the collimating lens design must simply apply a
phase delay equal in magnitude though in the opposite direction to
the input wavefront. This is achieved by first computing .phi.(x,y)
(see, FIG. 5), then wrapping .phi.(x,y) from 0 to 2.pi.
radians:
.phi.'(x,y)=arg [0(x,y)]=arg[exp(i*.phi.(x,y))]
[0031] Where the argil operator returns the phase-angle of a
complex number 0(x,y). It may then necessary to quantize the phase
profile .phi.'(x,y). This depends on the manufacturing method.
Diffractive optics can be formed using many techniques, and in this
case a lithographic etching process is assumed which requires the
surface to be quantized into discrete phase levels. This phase
quantization is not to be considered a limitation or requirement as
manufacturing methods exist that permit a continuous phase profile;
however for completeness of the example, a quantization step is
provided here. Four quantized phase levels are defined between 0
and
6 4 .pi. , ##EQU00001##
more generally, the range of quantized levels is between
0 .fwdarw. 2 .pi. ( N - 1 ) N . ##EQU00002##
The quantization of the phase profile .phi.'(x,y) into N levels is
defined as:
.phi..sub.N(x,y)=[.phi.'(x,y);N]
[0032] where the operator defines the quantization operation on a
phase profile. In this example, N=4 such that:
.phi..sub.4(x,y)=[.phi.'(x,y);4]
[0033] Note that any integer value for N may be selected. N=4 is
used herein so that the illustration is made more clear. In
general, as N increases, the efficiency of the optic (i.e., the
performance) increases, so it is beneficial to design optics with
more levels as long as the number of levels remains manufacturable
with respect to the diffractive surface.
[0034] Once the quantized phase profile of the lens is computed, it
is rotated about its origin to form a radially symmetric lens
surface. An example of such a surface is shown in FIG. 6.
[0035] Beam shaping optic design: The beam shaping function may be
designed to have an arbitrary effect on the output intensity
distribution of the incident light source. For example, it may be
useful to generate a high resolution grid of points in the output
field.
[0036] The method for optimization is based on an infinite point
source (i.e., collimated) illumination, for which there are many
known and published solutions in the literature including the
Gerchberg-Saxton algorithm (Gerchberg R. W and Saxton W. O., "A
Practical algorithm for the determination of phase from image
diffraction plane pictures," Optik (Stuttgart), 35, 237-246, 1972,
incorporated by reference), the first of a family of algorithms
referred to the art as an Iterative Fourier-Transform Algorithm
(IFTA), as well as alternative, global-search algorithms for
optimization. Whichever algorithm is used, the outcome is a phase
profile .chi.(x,y) which describes the necessary shape of the
wavefront (in phase angle) to produce the desired output intensity
distribution in the far-field.
[0037] In an example using the Gerchberg-Saxton algorithm, a beam
shaping optic phase profile .chi.(x,y) is generated. The optic
function, like the diffractive lens design described above, is in
the phase domain. By this it is understood that it has no physical
depth, but rather describes the required shape of the wavefront at
its output. The optic function in this example is to generate a
square top-hat profile. However, it can be any arbitrary output
distribution.
[0038] The optic function must be quantized from an infinite number
of levels (see, FIG. 7A) to a finite number of levels so that it is
compatible with the chosen manufacturing process. As in the example
above for the diffractive lens, the phase profile is quantized to
N=4 levels using the same quantization operator as described above.
FIG. 7B shows the quantized phase profile .chi..sub.4(x,y). It will
be noted that it is also possible to integrate the quantization
process within the IFTA in a so-called `soft-quantization` process
(see, for example, Wyrowski F., "Diffractive optical elements:
iterative calculation of quantized, blazed phase structures," JOSA
A, Optical Society of America, 7, 961-969, 1990, incorporated by
reference). So, in general, it should be clear that there are
alternative ways to include quantization into the beam-shaping
optic optimization.
[0039] In an embodiment, the quantized phase profile of FIG. 7B may
be repeated in a tile-like fashion to define an overall quantized
phase profile as shown in FIG. 7C.
[0040] Combining optics: As noted above, the diffractive lens
design and the beam shaper design are both in the phase domain.
[0041] The diffractive lens design and beam shaper design are
combined by summing together the respective quantized phase
profiles:
.rho.(x,y)=.phi.(x,y)+.chi.(x,y),
[0042] In particular, with N=4, the following describes the
operation:
.rho..sub.4(x,y)=.phi..sub.4(x,y)+.chi..sub.4(x,y),
[0043] This is a convenient representation as various methods for
physical realization of the phase delay in a single diffractive
surface are possible to form the diffractive optical element 116.
Examples of such physical realizations include a surface relief
pattern or a gradient-index (GRIN) material. A specific but not
limiting example using a surface relief pattern is provided
herein.
[0044] FIG. 8 illustrates the result of summing the phase profiles
with respect to FIGS. 6 and 7C.
[0045] The resulting phase profile .rho.(x,y) is then wrapped into
2.pi. radians:
.rho.'(x,y)=arg[exp(i*.rho.(x,y))]
[0046] For this example, the following is true:
.rho..sub.4'(x,y)=arg[exp(i*.rho..sub.4(x,y))]
[0047] FIG. 9 illustrates the result of wrapping the summed phase
profiles as in FIG. 8. One advantage of wrapping in this manner is
a compressing the vertical height of features of the physical
diffractive surface.
[0048] Using knowledge of the refractive index (n.sub.2) of the
material that the optic will be constructed from, the refractive
index of the immersive medium (typically air: n.sub.1=1) and the
wavelength of the light (.lamda.) for which the optic will be
optimized, the wrapped phase profile .rho.'(x,y) can then be
converted into a physical surface relief profile, S(x,y), using the
following relationship:
S ( x , y ) = .rho. ' ( x , y ) .lamda. 2 .pi. ( n 1 - n 2 ) ,
##EQU00003##
or
[0049] for quantized phase profiles:
S N ( x , y ) = .rho. N ' ( x , y ) .lamda. 2 .pi. ( n 1 - n 2 ) .
##EQU00004##
[0050] So, it follows in this example with N=4 that:
S 4 ( x , y ) = .rho. 4 ' ( x , y ) .lamda. 2 .pi. ( n 1 - n 2 ) .
##EQU00005##
[0051] Evaluation of the collimating and beam shaping diffractive
optical element 116: There are many sources of error that can
degrade the form factor of the diffractive optic with respect to
the initial design form. If magnitudes of these errors are set to
limits expected in manufacture, the error in the height of the
module (i.e., the depth error) is recognized as the key contributor
to degradation of optic performance. FIGS. 10A-10C illustrate
simulated results of the DOE 116 with a nominal depth error (i.e.,
less than or equal to 5%) in FIG. 10A, a 10% depth error in FIG.
10B and a 15% depth error in FIG. 10C. It will be noted that even
with a significant depth error, the DOE 116 fails to exhibit
concern with respect to zeroth order leakage (compare to FIG. 3).
Furthermore, even at a 50% depth error, the result remains
significantly homogenized, while experimentation shows that the
zeroth order leakage characteristic of the prior art would instead
begin to dominate when a 10% depth error is manifest (see, FIGS.
3B-3C).
[0052] The foregoing description has been provided by way of
exemplary and non-limiting examples of a full and informative
description of the exemplary embodiment of this invention. However,
various modifications and adaptations may become apparent to those
skilled in the relevant arts in view of the foregoing description,
when read in conjunction with the accompanying drawings and the
appended claims. However, all such and similar modifications of the
teachings of this invention will still fall within the scope of
this invention as defined in the appended claims.
* * * * *