Skip to main content

Full text of "Holographic Optical Elements as Scanning Lidar Telescopes"

See other formats


Popular Summary 
Holographic Optical Elements as Scanning Lidar Telescopes 

Geary K. Schwemmer 

Laboratory for Atmospheres, NASA Goddard Space Flight Center, Greenbelt, Maryland, 20771 

Richard D. Rallison 

Ralcon, Inc., Paradise, UT 84328 

Thomas D. Wilkerson 

Utah State University, Logan, Utah 84322-4405 

David V. Guerra 

Saint Anselm College, Manchester, New Hampshire 03102. 

NASA has developed new telescope systems that use holograms instead of lenses or 
mirrors. These holographic telescopes were developed in an effort to reduce the size, 
weight and cost of laser instruments used to measure atmospheric properties like 
temperature and wind. Several hologram designs were made and tested in the laboratory. 
Two were incorporated into laser remote sensing systems that are used on the ground and 
in airplanes to study the atmosphere. Our goal is to enable the development of spacebome 
scanning laser remote sensors for Earth science applications using the lightest and least 
costly technologies available. 



Holographic Optical Elements as Scanning Lidar Telescopes 

Geary K. Schwemmer 

Laboratory for Atmospheres, NASA Goddard Space Flight Center, Greenbelt, Maryland, 20771 

Richard D. Rallison 

Ralcon, Inc., Paradise, UT 84328 

Thomas D. Wilkerson 

Utah State University, Logan, Utah 84322-4405 

David V. Guerra 

Saint Anselm College, Manchester, New Hampshire 03102. 

We have investigated and developed the use of holographic optical elements (HOE) and 
holographic transmission gratings for scanning lidar telescopes. By rotating a flat HOE in 
its own plane with the focal spot on the rotation axis, a very simple and compact conical 
scanning telescope is possible. We developed and tested transmission and reflection HOEs 
for use with the first three harmonics of Nd:YAG lasers, and designed, built, and tested 
two lidar systems based on this technology. 

Introduction 

Lidar is making significant contributions to those Earth sciences requiring remote measurements 
of atmospheric and surface parameters from ground-based, airborne, and spacebome platforms. 
Scanning provides the means for increasing topographical coverage in airborne laser altimeters, 
and for generating three-dimensional data sets using ground-based and airborne atmospheric 

1 



lidar systems. Scanning will enable high-density global coverage for observing atmospheric 
dynamic parameters from space, such as cloud and aerosol structure, temperature, and humidity. 
Of great interest to the atmospheric science community is the possibility for frequent, high 
vertical resolution, global atmospheric wind profiles. Spacebome Doppler lidar is currently 
deemed the most feasible means of obtaining these measurements, and a scanning, pointing, or 
multiple-look-angle telescope system is required to retrieve full horizontal wind vectors. 
However, most of the scanning atmospheric lidar systems conceived to date have been too heavy 
and costly to develop into spacebome versions. Improvements in efficiency, size and weight are 
required of all technologies involved in lidar remote sensing in order to realize the advantages 
over passive sensors that lidar offers, such as high accuracy and vertical resolution. Large 
improvements may come from developing more efficient laser transmitters, which for 
atmospheric lidars are usually the heaviest components when the requisite power supplies and 
radiators are included. The next largest or most massive component is the receiver telescope. 
Typically, lidars require a large collecting aperture to maximize the laser backscatter signal, and 
a narrow field-of-view (FOV) to limit the amount of background radiation reaching the detector. 
In order to scan a conventional lidar system, the entire telescope assembly is steered, or a large 
flat steering mirror is placed before the telescope to point the FOV in different directions. Focal- 
plane scanning approaches have been used in altimetry lidars to scan over several degrees, but 
are generally not viable for atmospheric lidar requiring scans over 90 degrees or more. NASA is 
investigating a number of innovative telescope technologies, including deployable optics, ultra- 
lightweight materials, and the use of diffractive optical technologies for various applications. A 
large reduction in instrument weight can come from utilizing these new optical receiver 



technologies individually or in combination, one of which is the holographic scanning telescope 
described here. 

We have developed and experimentally investigated the concept of a scanning telescope 
using a holographic optical element (HOE) in order to help reduce the satellite resources and 
costs needed for a large, orbiting laser remote sensing instrument for measuring atmospheric 
parameters: e.g. wind, temperature, and humidity profiles'"'*. By rotating an HOE in its own 
plane, we create a conical scan pattern with a minimum of mechanical and electrical 
requirements. Used in this manner, the holographic scanning telescope offers advantages over a 
conventionally scanned telescope in that it reduces complexity and number of parts by 
substituting a single HOE for the receiver telescope primary optic and the scan mirror. HOEs can 
be reproduced at low cost compared to large aperture reflectors or refractors. We also 
investigated using holographic gratings as scanners in conjunction with static HOEs and 
conventional telescopes, and find special care must be taken to avoid spurious signals from the 
zero and minus-one orders when these optics are also used to transmit the laser beam. 

Review of Scanning Lidars 

Scanning lidar systems have been used for a number of applications, using a variety of scanning 
techniques. Conflicting requirements often arise that tend to drive the scanning system design, 
with cost usually being a major factor. Most atmospheric lidars utilize large aperture telescopes, 
which are usually more cost effective and reliable at gaining signal than the use of larger laser 
transmitters. But the instantaneous FOV (IFOV) of these systems is often very small, usually less 
than 1 mrad, in order to reduce daytime solar background. There are in general three ways of 
scanning a telescope with conventional optics. One is to mount the telescope and associated 
transceiver optics on a scanning mount. Such mounts are relatively large and expensive in order 



to accommodate the mass and inertia of the telescope assembly. Astronomical telescopes and 
tracking mounts are examples of this type. An early example of a lidar employing this type of 
scanning is the Large Atmospheric Multi-Purpose lidar^. More recently is the design proposed 
for what would have been the first scanning spacebome lidar, the Atmospheric Lidar 
Instrument , but that has since been changed to a non-scanning system. 

The second type of scanning telescope utilizes one or more large flat scanning optics in front 
of the telescope aperture. An example using one mirror mounted on a single-axis scanning mount 
is the Goddard Scanning Raman Lidar^. Another example of this type is the lidar mounted on the 
NASA DC-8 during the SUCCESS campaign* in which the scan mirror was mounted within a 
pod on the outside of the aircraft's fuselage. Hwang's lidar terrain mapper system^ also uses a 
single rotating flat scan mirror to achieve cross-track coverage from an airborne platform. Some 
systems are designed with a single flat mirror on a two-axis mount. This allows for somewhat 
more flexibility, including 3-dimensional volume imaging at the expense of a more limited scan- 
angle range, and requires a somewhat larger flat mirror. The Large Aperture Scanning Airborne 
Lidar'*' uses a two-axis scanning mirror in front of a 55 cm diameter telescope situated in the 
bomb bay of the NASA P-3B aircraft. The telescope is mounted horizontally, pointing aft at the 
scan mirror. The scan mirror is nominally tipped at 45 degrees and rotates ± 45 degrees on a 
shaft aligned with the telescope optic axis to effect cross-track scanning. A big disadvantage of 
this arrangement is the large scanning flat mirror is not only heavy, but requires a large motor 
and gearbox to achieve the torque required to constantly accelerate and de-accelerate the mirror. 
In addition, since the scanning mirror is within the confines of the fuselage, a large open area is 
required to permit an unobstructed 90-degree cross-track scan, achieved in the P-3B by removing 
the bomb bay doors. Other examples of lidars using single 2-axis scanning flats include those of 



Irish and Lillycrop", Hawley et. al.'^, and Bennett et. al.'^. Sometimes refractive rotating 
wedges are used instead of mirrors to generate a conical scan in a more compact design. The 
Wind Infrared Doppler Lidar''* is one example of a system using a rotating germanium wedge to 
scan its infrared transceiver telescope. Similarly, the Multi-center Airborne Coherent 
Atmospheric Wind Sensor'^ used two rotating wedges to steer the IFOV over any direction 
within a limited total angular range. The majority of 2-axis scanning lidars utilize two flat 
mirrors each on its own single-axis mount, in an azimuth - elevation scan configuration. This 
usually allows for complete hemispherical coverage in ground-based systems, but is awkward 
and expensive to incorporate into airborne systems and would be prohibitively large and heavy 
for spacebome use. Example lidars with two scan mirrors are the University of Wisconsin's 
Volume Imaging Lidar'^, the system of Hooper and Martin'^, the Raman water vapor lidar of 
Eichinger et. al.' , and the Goddard Lidar Observatory for Winds'^. Occasionally rotating 
polygon mirrors are used for rapid scanning in one axis, but these systems are limited in size to 
smaller apertures and are usually applied in terrain mapping or other hard-target lidars such as 
the one described by Chen and Ni^*^. 

The third major category of scanning lidar telescope using geometric (conventional) optics 
involves focal-plane scanning mechanisms. These systems utilize either a small scanning mirror 
in the focal plane, or an array of detectors combined with separate transmitter beam steering 
optics. This type of scanning requires a telescope with a total FOV as wide as the angular extent 
of the scan. This appreciably adds to the cost of a large telescope when the scan angles are 
greater than a few degrees and is generally limited to altimeters and similar hard-target lidars. 
The Laser Vegetation Imaging Sensor^' is one example of an airborne scanning lidar that uses a 



small oscillating galvanometer scan mirror in the focal plane, coupled with a wide FOV 
telescope, to achieve ± 7.5 degree cross track scanning. 

Holographic Optical Elements 

An HOE can be thought of as a hologram of a point. It is constructed using the fringe pattern 
generated by the interference between two beams of mutually coherent monochromatic light. 
(Fig. la) One beam (called the object beam) is from a point source and has spherical wavefront. 
The other (called the reference beam) is collimated, having a planar wavefront. After exposing 
and processing is complete, one can "play back" or reconstruct the recorded image of the point 
source by illuminating the HOE with light from a collimated or quasi-collimated source 
conjugate (in the opposite direction) to the original reference beam (Fig. lb). The fact that it can 
be illuminated simultaneously by light from many such reference beams such as from a distant 
extended source or object, each reconstructing a corresponding point in the image plane, gives 
the HOE lens-like imaging properties. We refer the reader to Close^^ for a a basic treatment of 
HOE formation, imaging and dispersion properties, including a comparison with conventional 
imaging optics. In addition to acting as a telescope's light collecting objective, the HOE can also 
be used to transmit the laser beam in a lidar system, coUimating and steering it at the same time. 

The use of HOEs for scanning lasers is not new, nor is the idea of using one as a lidar 
telescope. Rallison^^ conceived of using a static HOE as a spectrally discriminating collecting 
telescope in a laser range finder. Gilbreath et. al.^'* discussed using HOEs as lightweight 
transceiver optics for collimating and correcting astigmatism in diode laser transmitters for 
spacecraft optical communication systems. The use of HOEs in laser beam scanners for optical 
bar code readers is commonplace, with many patents existing in this area. However, most such 
previous rotating scanners attempt to straighten out the path of the scanned beam, which would 



otherwise describe a circle. These systems use small pie-shaped segments of the scanner, each 
containing a separate HOE to scan the beam at different angles. This makes for very inefficient 
use of the available aperture, which is the opposite design feature required by typical lidar 
systems, where the weakness of the return signals necessitates the use of large photon collecting 
apertures. 

The first HOE we designed and constructed to use with a conical scanning lidar was a 40 cm 
reflection hologram for use with a 532 nm wavelength laser\ It was tested in the laboratory and 
in a breadboard lidar system before being incorporated into what is now the Prototype 
Holographic Atmospheric Scanner for Environmental Remote Sensing (PHASERS)^^, and 
discussed in more detail later in this paper. Encouraged by the successful demonstration of our 
concept, we proceeded to develop HOEs for use with other laser wavelengths^^' '^^ and 
applications. This led to development of the Holographic Airborne Rotating Lidar Instrument 
Experiment (HARLIE) based on a transmission HOE for use with the 1064 nm wavelength of 
the Nd:YAG laser^ . The use of a transmission HOE allows the system to scan over wide angles 
through a similarly sized window in the aircraft. We are currently developing holographic 
telescopes for use with UV (355 nm) wavelengths, 1 -meter diameter apertures, and as dispersive 
optics for Raman lidar applications . We are also investigating further reduction in mass by 
eliminating all moving parts ' using angle multiplexed HOEs. 

In the remainder of this manuscript we describe the manufacturing and optical testing of the 
HOEs used in our work. We give a short description of the two holographic scanning lidars we 
have developed. We then discuss several practical aspects of using HOEs in lidar receivers 
before concluding. 

Fabrication of HOEs 



For our purposes HOEs can be divided into two categories: reflection HOEs, in which the 
incident and diffracted light are on the same side of the HOE, and transmission HOEs, in which 
incident and diffracted Hght are on opposite sides of the film. We investigated both reflection 
and transmission HOEs in this program. 

Reflection HOEs have the advantage of being able to utilize opaque substrates to support the 
holographic film or pattern, which opens the possibility of being able to use ultra-lightweight 
materials such as graphite epoxies and similar composites. Volume phase reflection HOEs must 
be produced on transmissive optical substrates and may be applied to opaque substrates using 
film transfer techniques, although these techniques have not been developed for large (~>10 cm) 
holograms. However, reflection HOEs can also be produced as surface relief holograms using 
reactive ion etching or other techniques directly to an opaque substrate. Reflective coatings can 
be applied to boost the diffraction efficiency. 

Transmission HOEs must be applied to optically transmissive substrates such as glass. They 
are well suited for airborne systems in which one is restricted to having the instrument view 
through a pressure-sealed window. Placing the receiver HOE very close to the window 
minimizes the size of window needed to accommodate the wide scan angle. The Bragg planes in 
reflection HOEs tend to be oriented at oblique angles relative to the optic axis, whereas in 
transmission HOEs they tend to be oriented at acute angles with respect to the optic axis (which 
is normal to the film in our examples). Due to this aspect of the fringe-plane structure in 
transmission HOEs, it is also easier to reduce aberrations to the image quality, allowing one to 
use a smaller FOV to help decrease daytime solar background levels in lidar telescope 
applications. 



We first produce a master HOE from which working copies are made. The master is 
produced by exposing a glass plate coated with a film of dichromated gelatin (DCG) emulsion to 
two mutually coherent laser beams as illustrated in Fig. la. Molecular cross-links are formed in 
the gelatin upon exposure to light, thus the interference fringes are registered in the film as 
variations in hardness and index of refraction. Post-exposure processing is performed on the 
film by alternately immersing the film in water to remove the photosensitive dichromate from the 
gelatin and hot isopropyl alcohol to remove the water. The resulting hologram is then relatively 
free of absorption. The HOE is then oven dried and tuned if necessary to adjust the peak 
response wavelength by adjusting the water content of one of the alcohol processing baths and 
reprocessing the plate. Adjusting the water content causes the film to shrink or swell in 
thickness, thereby changing the tilt and modulation of the fringe planes and hence the center 
wavelength and efficiency at the correct diffraction angle. The HOE focal spot size is checked, 
and if within specification the film is hermetically sealed with a cover glass cemented to the film 
side of the substrate and sealed around the edges with UV curing epoxy. The spot size is 
rechecked, and if it is still acceptable, an anti-reflection (AR) coated cover glass is epoxied to the 
original uncoated cover glass. Since many trials are often required to produce a satisfactory 
master HOE, the AR coated glass is not used throughout this process due to cost considerations. 

The development of the first HOE for practical atmospheric lidar applications faced the 
challenge of making a large (40 cm) diameter optic having both high diffraction efficiency and a 
small focal spot size. This had to be done using a relatively short focal length in order to design a 
receiver system compact enough to be competitive with conventional Cassegrain telescopes 
typically used in lidar receivers. The final design choice was an f/3.2 reflection HOE for use at 
532 nm. 



The second major step to eventual spacebome use was developing an airborne lidar system to 
not only demonstrate the HOE technology, but would also have scientific utility. We wanted to 
work at the Nd:YAG laser fundamental at 1064.7 nm to measure atmospheric aerosol 
backscatter. DCG is non-absorbing at this wavelength. The only suitable laser available for 
exposing this hologram was a 488 nm Argon ion laser. At the same time we wanted to decrease 
the focal spot size in order to be able to utilize the HOE in a lidar system having daytime 
capability with the small laser pulse energies produced by a diode-pumped Nd:YAG laser. 
Spherical and other aberrations can increase dramatically when a hologram is played back at a 
wavelength well removed from the exposure wavelength. A number of approaches were 
considered and tried in developing an HOE that would work well at 1064 nm when exposed at 
488 nm. The final approach that met with great success was the method of "time-reverse ray 
tracing"^^. Once this technique was perfected, the designs for several transmission HOEs were 
developed and built in rapid fashion for a variety of lidar applications that would employ 770, 
1046, 832, 523, and 532 nm light. 

In general, the HOEs were fabricated on Solarphire or B270 float glass that was spin coated 
with 10-micron thick films of DCG. Several copies of each design were exposed, processed and 
tested. The best device in each case, in terms of focal spot size, was capped with an AR coated 
Solarphire cover and used as a master from which contact copies were made. The masters were 
made to diffract about 50% when illuminated with 488 nm light. Contact copies were produced 
by placing the master HOE in close contact over a glass substrate coated with a fresh layer of 
unexposed film and its intended cover glass. Index matching xylene is forced and held between 
the three pieces by capillary action. A laser beam sheet was produced with cylindrical and 
spherical lenses, then reflected 90 degrees off a long, narrow flat mirror on a mount moving 



10 



parallel to the HOE to scan the master /copy film (|Hi2). The scanning laser light passes 
through the master HOE to the unexposed film. A portion of the light is diffracted into the first 
order, forming a collimated sheet traveling at an angle with respect to the undiffracted beam with 
which it interferes to produce the fringes that will expose the copy film. Because the distances 
between the interfering wavefronts are short, the coherence length of the laser used to make the 
copies need not be as great as that making the original. And because the laser light is only 
formed into a sheet rather than a fully collimated beam, the intensity per unit area is greatly 
increased, greatly decreasing the required exposure time for any given portion of the film. For 
these reasons, the scanning mirror can be moved at a moderate pace and does not require 
interferometric stability. 

Several of the HOEs and holographic grating scanners made during the course of this 
research are pictured in Fig. 3. Clockwise from the left foreground: A 25 cm transmission grating 
for use at 1046 nm, a 40 cm transmission HOE for 1064 nm, having a 100 cm focal length and 
45 degree diffraction angle; a 40 cm tiled mosaic transmission grating optimized for 770 nm at 
45 degrees. This optic was made in four exposures (one in each quadrant) to demonstrate one 
method of scaling the manufacturing process to meter sizes. The hole in the center is for 
transmitting a high power Q-switched YAG laser beam, which would be steered in the proper 
direction with a prism mounted in the hole and rotating with the grating. To the right is a 30 cm 
reflection HOE for 532 nm light at 45 degrees, and in the front center is a 20 cm transmission 
grating similar to item #1. Notice the diffracted image of the coffee mug just to the right of the 
mug placed behind the hologram. In the lidar, we transmit the outgoing laser light through a 
concave lens f-matched to the HOE, which then collimates the light while diffracting it at an 
angle to the substrate. Laser light backscattered by the atmosphere acts as the reconstruction 



11 



beam and is focused on the center normal to the HOE. Spinning the HOE about the center 
normal axis generates a conical scan with the transmitted light and the receiver FOV (Fig. 4). 
Components from the field stop to the detector remain fixed so no slip rings are required. This 
makes for a simple compact design. 

Optical Performance 

Being diffractive, HOEs are spectrally dispersive. The dispersion is determined by the surface 
grating defined by the intersection of the Bragg planes with the surface of the film and is given 
by the well known grating equation: 

sin a + sin yff = mA/d, Eq. 1 

where a is the input angle, P is the output angle, m the diffraction order number, A the 
wavelength of the light used to "playback" the hologram, and d is the surface spacing of the 
Bragg-plane fringes. To rotate the HOE in its own plane and keep the focus on a fixed point, the 
angle ^ must be 0°. Spectral dispersion helps to filter out background light, since the light at 
undesired wavelengths is dispersed in the focal plane and will fail to enter the field stop aperture. 
However, light from different parts of the sky at other wavelengths (within the bandpass of the 
HOE) will be diffracted into the field stop. 

The spectral bandwidth AA for a transmission HOE designed for use at a wavelength A is 

AA^Ad/Ti&x\(l), Eq. 2 



and for a reflection HOE is 



AA-^Ad/T, Eq. 3 



12 



where = | a„+^n I /2 is the diffraction half-angle, the subscript n denotes angles inside the film, 
and T is the thickness of the film. 

The expected efficiency ;;, of a transmission HOE is given by 



7] ~ sm\^AnT / Acos^], Eq. 4 



and for a reflection HOE by 



Tj ~ tanh^[;rzlnr//lcos<z>], Eq. 5 



where An is the peak-to-peak index modulation (the difference between the extremes in index of 
refraction values in the fringes.^^ 

For example, using i=532 nm, a typical film thickness of 10 ^m, refractive index 1.27, 
An=0.03, a=45°, and y5=0°, a transmission HOE will have a peak diffraction efficiency of 89%, 
and a bandwidth of about 132 nm. Similarly, a reflection HOE with similar parameters will have 
a peak efficiency of 53% and a bandwidth of 40 nm. The amount of light diffracted as a function 
of wavelength was measured for the first 532 nm reflection HOE, using a high pressure Mercury 
arc lamp at the focus of a collimating parabolic mirror to illuminate the HOE (Fig. 5). The peak 
response (uncorrected for lamp output) occurs around 528 nm and the spectral bandwidth 
(FWHM) is about 46 nm, within 15% of the expected value. Frequently, bandwidths calculated 
using Eqs. 2 and 3 do not agree this well with actual bandwidths, because of gradients and chirp 
in the actual diffractive structure in the HOE. Depending on how a plate is processed, it may play 
back as if it had 40-80% of the original film thickness because most of the modulation is in the 
top few microns and not evenly distributed throughout. 



13 



Table 1 lists the various HOEs and gratings that were designed, fabricated, and tested as part 
of this program. Item #1R is the PHASERS reflection HOE made in 1989, described earlier in 
this paper. Item #2 is a prototype to the HARLIE HOE. Items #3-5 are transmission gratings 
conceived for use with existing lidars having conventional telescopes. Using improved 
fabrication techniques developed during this program, #6R was made in an attempt to improve 
on the efficiency and angular resolution of #1R. Its incident angle was designed to be 43 degrees 
to match the actual diffraction angle of #1R, in order to replace it in the PHASERS system 
without having to make any mechanical changes to the lidar. Items 7a, 7b, 8a, and 8b are 
designed to be used in pairs for Nd:YLF-based terrain mapping lidars. The a's are transmission 
gratings that rotate to perform the scanning function, and the b's are transmission HOEs with the 
collimated beam normal to the optic and the focus off -normal. 

We tested the HOEs at their design wavelengths for focal spot size, diffraction angle, 
efficiency, and focal length using a horizontal collimated beam of laser light at the appropriate 
playback wavelength, expanded to fill the diameter of the HOE (Fig.6). To correctly orient an 
HOE to the collimated beam, first the tilt about a horizontal axis along a diameter of the HOE is 
adjusted so that the specular reflection from its front surface will remain in the horizontal plane. 
The front surface is identified during manufacturing as the side that a collimated light source 
should impinge upon to create a focused spot by diffraction. The HOE is then aligned about a 
vertical rotation axis (also a diameter of the HOE) while monitoring the focal spot with a CCD 
camera, to give the minimum spot size. The HOE is then rotated in its own plane to bring the 
focal spot into the horizontal plane. Then the vertical rotation angle is readjusted to minimize the 
spot size. This establishes the proper incidence angle of the collimated beam with the HOE and 
the plane of diffraction. We usually find small differences between the incident angle that 



14 



produces the smallest focal spot and the incident angle that produces the highest diffraction 
efficiency, but this is usually less than a degree. It is a matter of judgment exactly which angle to 
use. The spot size is more sensitive than efficiency to departures from the optimum incidence 
angle, efficiency only changing by 1 or 2 percent over a few degrees. So the optimum angle is 
the one that produces the smallest spot (Fig. 7). The plane gratings were tested only for 
diffraction efficiency and diffraction angle. 

Diffraction Efficiency 

To test the diffraction efficiency of each of the holograms we measured the total energy 
incident on the HOE and the energy in the first-order diffracted spot. The measurement of the 
light incident on the HOE was made using a Fresnel lens to collect the light from over the full 
aperture and focus it onto a 1 cm power meter, as shown in Fig. 6. The same Fresnel lens and 
meter were then placed on the opposite side of the HOE to measure the diffracted power. We 
divided the diffracted power measurement by the incident power measurement to calculate the 
first-order diffraction efficiency. We also measured the zero-order (undiffracted) transmission in 
this fashion. The test data for the HOEs listed in Table 1 are recorded in Table 2. The 
percentages in the zero and first orders do not add up to 100 % because some light is lost to 
scattering, absorption, Fresnel reflections, and other diffraction orders. Item #7a was also tested 
at 904 nm to see how it might perform if used with a diode laser altimeter system at that 
wavelength. 

The focal lengths were measured with calipers and a ruler from the focal point to the center 
point of the surface of each HOE. 



15 



The diffraction angle for each grating was measured directly, by retro-reflecting the first 
order light with a large, flat mirror, and then measuring the difference in angle between the flat 
mirror and the HOE with a theodolite (Fig. 8). 

Different methods were used to measure the diffraction angles of the HOEs. If the HOE was 
designed with the collimated beam off-normal, the incident angle was measured using the 
theodolite in a manner similar to the gratings. If the HOE was designed with the collimated beam 
normal and the focus off-normal, the diffraction angle of the focus-side optical axis was 
calculated geometrically using measurements of the focal length and its displacement from the 
central normal ray. 

Spot Size 

The focal spot size of each HOE was measured with a CCD camera set in the focal plane of the 
HOE's. The focal spots were usually slightly astigmatic, with the better HOEs having astigmatic 
differences of -1-2 mm. The CCD imaging system software had provisions to calculate the 
encircled energy as a fraction of the total energy falling on the detector (after a background 
subtraction) for any size circle or ellipse drawn on the image (Fig. 9). For round looking spots, 
we found the circle, centered on the energy centroid of the focal spot, that contained about 86.5% 
of the total energy. If the focal spot resembled an ellipse, then the FWHM and 1/e^ points were 
measured for each axis of the ellipse. The encircled energy versus the diameter of such focal 
plane apertures are plotted for HOEs #1R and 2 in Figs. 10 & 11. 

It may seem that by using float glass in the construction of the HOEs that the focal spot sizes 
would be limited by the aberrations induced from the glass. However, the performance is limited 
by random non-uniformities in the bulk index of the gelatin induced by non-uniformities in the 
chemical processing used to create the index modulation. The process liquids always leave a trail 

16 



behind as they run off the film following removal from the processing vats. At the surface of the 
gel these trails are seen as very small surface deformations. Each low spot has under it a higher 
density of gel than each high spot, so that optical path thickness is constant. Thus, planar wave 
fronts are not distorted due to refraction by the uneven surface. If a substance with uniform 
density, like epoxy, is applied to that surface and allowed to fill in all those low spots so that the 
exit surface is now level, then the time to transit from the bottom of a low spot to the new exit 
surface has been increased relative to the time to transit from a nearby high spot to the same 
surface. The result is refraction-induced distortions in beam steering. This effect is believed to be 
the current major source of wavefront errors introduced in light diffracted by the HOE or grating. 
It typically introduces between 100 and 200 pirad of aberration to the wavefronts. 

To minimize this effect in gelatin the processing needs to be improved so that it is more 
uniform and creates fewer high and low density regions with associated high and low elevations 
at the surface as well as other volumetric distortions of the fringe structure. Then a cover glass or 
a layer of epoxy will not create spot enlargement and we expect to see 50 /xrad spots, even with a 
wave or two of error in the recording optics and cover glass. 

Other sources of wavefront errors include residual spherical aberration and astigmatism not 
completely corrected in the exposure optical design. 

Prototype Lidar Systems based on HOE telescopes 

We designed and constructed a simple lidar to test the concept of a scanning holographic 
telescope system based on the green reflection HOE #1R. The HOE is the disk at the far end of 
the bench at the bottom of Fig. 12. The laser transmitter (ImJ @ 532 nm, 20 pps) is the dark box 
located on the left, emitting a beam through a diverging lens and a beam tube to a 45-degree 
turning mirror that directs the beam down and normal to the center of the HOE. The HOE 

17 



collimates the laser beam while diffracting it up at a 43.5 degree angle to the vertical. At this 
point the beam is about 4 cm in diameter. The backscattered radiation is collected by the entire 
HOE aperture and focused to a 2.0 mm field stop located in the dark cylinder supported on the 
tripod directly above the HOE. The central portion of the HOE that is used to transmit the laser 
beam is obscured from the detector by the beam turning mirror mount and housing located on the 
BHI assembly just above the HOE. 

The HOE is mounted on a manual rotation stage in order to scan the system. A photon 
counting photomultiplier detector is mounted in the light colored cylinder directly above the field 
stop with a 10 nm interference filter between them to help cut down on stray background light. 
The concept was successfully tested in nighttime operation in this configuration^^. 

In 1995, a motorized turntable was installed in the system and it was moved to the roof of the 
physics building at Western Maryland College (since renamed McDaniels College) in 
Westminster, Maryland. The system was dubbed PHASERS and has been the subject of 
undergraduate research projects^^. 

PHASERS was relocated to Saint Anselm College in Manchester, New Hampshire in 1998 
and outfitted with a sturdier structure and baffles (|||m|3) It continues to be used to introduce 
undergraduates to the field of lidar remote sensing. 

The second lidar to use a scanning HOE telescope is HARLEE. HARLIE is a technology 
demonstration to test the utility of using holographic scanning receivers in lidar systems at the 
1064 nm Nd:YAG wavelength and in an aircraft environment. Built as an aerosol backscatter 
lidar system, HARLIE uses a 40 cm diameter by 1 cm thick transmission HOE as the laser 
collimating optic as well as the receiver primary optic. It has a 45 degree diffraction angle and a 
1 meter focal length normal to its surface. It rotates continuously at rates up to 30 rpm, and can 



18 



also operate in step and stare or static modes. Improvements to the HOE design and 
manufacturing enabled us to obtain a 160 /irad focal spot encircling 86.5% of the diffracted 
energy. The higher angular resolution greatly improved daytime performance over PHASERS. 
The focus is placed on a 200 /xm, 0.22 NA fiber optic field stop that delivers the backscattered 
light to the aft optics package. The aft optics contains a collimating lens, a 500 pm interference 
filter, a focusing lens, and a Geiger-mode avalanche photodiode. The Q-switched transmitter is a 
CW-pumped Nd:YAG laser capable of delivering over 20 w output, pulsed or CW. In HARLIE 
we run it at a 5 KHz repetition rate with 400 /iJ pulse energy. The beam is expanded using a -61 
mm f. 1. lens before being transmitted through the center of the HOE, which collimates the beam 
to 70 jLirad divergence x 20 nmi diameter. The angular extent of the transmitted beam is smaller 
than the full-aperture FOV because the small portion of the HOE illuminated by the outgoing 
laser beam introduces fewer wavefront errors. 

Figure 14 is a photograph of the HARLIE transceiver assembly mounted on its transportation 
dolly and electronics rack. The lidar can also operate on the ground with the transceiver directed 
in any of 8 elevation positions spaced 45° apart. In this figure the system is pointed up, so the 
HOE appears on top, mounted in a large ring gear and ball bearing assembly. The laser is 
mounted on the bottom and the beam is directed into the box and onto the HOE rotation axis, 
exiting through the center of the HOE. The electronics rack contains the data system, the laser 
power supply and chiller, the scan motor controller, a GPS receiver, and an aircraft inertial 
navigation system interface. The entire transmitter/receiver package can be placed within 
centimeters of an aircraft instrument window. A 52 cm clear aperture window allows for an 
unobstructed view in all directions around the conical scan (Fig. 15). 



19 



Figure 16 is a sample of ground-based data taken with HARLIE in the scanning mode. The 
conical surface of the HARLIE scan is projected in an isometric view out to 5 km altitude and 
horizontal range. Increasing levels of backscatter are revealed as darker shades of gray. There is 
a thin stratus cloud layer at 2 km altitude and multiple thicker cloud layers between 3 and 5 km 
altitude, with virga streaming from the layer near 4 km altitude in the lower right hand portion of 
the cone. This single 36-second scan of data was taken on 9 June 1999 at Saint Anselm College 
in Manchester, New Hampshire during the HOLO-2 campaign^'*' ^^. It reveals the rich details and 
atmospheric aerosol structures that a scanning lidar system is capable of detecting. Sequential 
series of such scans can be analyzed for aerosol and cloud motions that can be used to infer 
atmospheric wind vectors^^' ^''. Examples of time-lapse visualizations and other HARLIE data 
are available on the HARLIE web site^*. 

Practical Considerations 

After extensive testing and field use of various configurations of holographic gratings and HOEs 
for lidar applications as the primary transceiver optic, we have some additional insights into what 
works and what doesn't work, and what may work better with improvements in design and 
construction. 

Note that with either HOE type the collimated beam cross-sectional area (transmitted and 
received) is an ellipse, with the effective lidar receiver collection area reduced from the actual 
HOE area by the cosine of the diffraction angle. 

We have found that the HOEs constructed in the manner described in Section 3 work without 
any ill effects when used with moderate power, low energy Q-switched lasers. The HARLIE 
laser typically produces 50 nsec, 400 ^J pulses at 5 KHz rep-rate, and illuminates the central 2 
cm diameter of the HOE for an average energy density of 130 |aJ/cm^ and average power density 

20 



of 640 mW/ cm^. Similarly, the green PHASERS laser (1 mJ, 20 ns, 2 kHz) has been used 
without any apparent degradation in the HOE performance over time. Even several years in an 
uncontrolled (but enclosed) environment and many hours of exposure to direct sunlight have not 
had any noticeable effect on its performance. However, when a small holographic grating was 
illuminated with the unexpanded output of a large pulse (600 mJ, 10 nsec) of a 532 nm Q- 
switched Nd:YAG laser having an energy density of ~1 J/cm^ , the absorption from the traces of 
residual dichromate and the epoxy used to encapsulate the HOE with a cover glass was sufficient 
to cause catastrophic optical damage. 

Boresight alignment between the laser and the HOE is essentially the same as for a lidar 
using a conventional telescope. For a coaxial or parallel transmitter-receiver alignment, one 
wishes to make the laser appear to emanate from the field stop when viewed looking back into 
the lidar from infinity. If the field stop is not positioned exactly on the rotation axis, the scan will 
still describe a cone, but the axis will be tilted with respect to the HOE rotation axis according to 
Eq. 3. 

One expects to have some polarization dependence of the diffraction efficiency in an HOE. 
In order to match the diffraction efficiency of s and p polarized light, one needs to control the 
average index and the index modulation values during the HOE manufacturing process . One 
has to sacrifice peak efficiency for either polarization in order to achieve equal polarization at the 
design wavelength and diffraction angle. The point at where both polarizations are equal is about 
85%. It is very difficult to control the manufacturing process parameters accurately enough to 
exactly match the efficiency for both polarizations. Typically there will be a few-percent 
difference between them. And because the diffraction angles vary over the surface of the HOE, 
there is a modest dependence of the diffraction efficiency over different areas of the HOE. Using 



21 



a linearly polarized laser, we have found the transmitted laser energy for the HARLEE HOE to 
vary with scan angle in a sinusoidal fashion by 5 % peak-to-peak, at half the scan period. This 
effect is minimized by circularizing the polarization of the laser transmitter. 

Gelatin is an elastic material and conforms to any substrate to which it is applied. When the 
substrate changes size due to temperature changes, the holographic film will follow. The primary 
effect is a change in the surface grating pitch on each surface of the film. This surface grating is 
what determines the optical geometry, i.e. the focal length and the diffraction angle will both 
change in direct proportion to the expansion rate of the substrate. A typical thermal expansion 
coefficient for glass of -10"^ yields a change in diffracted angle of about -15 |j,rad/K and a 
change in focal length of -10 nm/K for the HOEs we used in our lidars. The latter is of little 
consequence given the depth of field of an f/2.5 lens. The diffraction angle change will basically 
change the cone angle of the scan pattern. As long as the transmitter and receiver field stops are 
built with an athermal mechanical design, boresight alignment should not be detrimentally 
affected. Barring any large thermal gradients across the HOE optic, the transmitted beam and the 
receiver FOV should track together. 

We also experimented with using holographic transmission plane gratings placed in front of a 
conventional telescope to generate the conical scan. This technique has one disadvantage over 
using an HOE to perform the scanning if the grating is also used to transmit the outgoing laser 
beam. The -2-5% of undiffracted zero-order light constitutes a second collimated transmitted 
beam which will generate its own return signal that will be collected and combined with the main 
return signal from the -i-l diffracted order (F||| 17a). On its return, the backscattered light does 
not meet the Bragg condition for diffraction in the grating and passes through relatively 
unattenuated. If there are any clouds in the beam path, their backscatter signals may be 



22 



comparable in magnitude to that of cloud-free regions and will be superimposed on the first 
order diffracted beam signal. To make matters worse, the two signals will have a different range- 
altitude relationship depending on the pointing direction and orientation of the lidar. This 
problem can be alleviated by using a baffle tube for transmitting the beam after it leaves the 
grating. However, this baffle will have to be attached to the rotating optic in order to steer with 
the laser beam, adding to the mechanical complexity of the system. 

This "cross-talk" is not a problem in the HOE based telescope. The zero-order transmitted 
beam continues to diverge after leaving the HOE. Atmospheric backscattered light from the 0- 
order transmitted beam fails to match the Bragg condition and is not diffracted to a focus in the 
telescope (Fig. 17b). It is also best to use a design in which second order is evanescent. For an 
HOE in which the focal side axis is normal to the optic, then the collimated side diffraction angle 
should be 30 degrees or more to eliminate higher diffraction orders. 

That any given HOE is effective only over a limited spectral band opens the possibility of 
wavelength multiplexing HOEs using various approaches. One can make a sandwich of 
individual HOEs, each diffracting its own specific wavelength while being transparent to the 
other wavelengths. Also, one can produce multiple holograms in a single film using multiple 
exposures. Alternatively, multiple HOEs may occupy separate areas of the substrate, although 
that would not make efficient use of the available aperture. In order to accomplish the focusing 
of different wavelengths to separate detectors that remain stationary as the HOE is rotated, the 
foci may be placed at different locations along the rotation axis, delivering light to optical fibers 
which will carry the light to detectors located elsewhere (Fig. 18a). Alternatively, the foci may 
be arranged off-axis at various locations in the focal plane of a grating/HOE system in which the 
functions of scanning and focusing reside in separate optics (Fig. 18b). Further reductions in 



23 



cross-talk are possible by offsetting the diffraction plane of each HOE in azimuth angle. This is 
also a technique for angle multiplexing several HOEs for use at the same wavelength^^' ^'. 

Special consideration must be given to multiplexing HOEs for simultaneously diffracting 
laser harmonic wavelengths, as the mth order of the /nth harmonic will diffract efficiently in the 
same direction as the first order of the fundamental wavelength for any given HOE. For example 
if one wished to utilize the 1^' and 3^^ harmonics of a Nd:YAG laser and have the HOE diffract 
each of the wavelengths to separate detectors, one might use the design in Fig. 19. A reflection 
HOE designed to diffract 355 nm light is placed on top of a transmission HOE designed for 1064 
nm light. The fringes of the UV HOE are too close together to create constructive interference at 
1064 nm, hence the IR radiation will pass through to the transmission HOE underneath and be 
focused on the opposite side of the assembly. 

For observing different wavelengths that are close enough to fall within the diffractive 
bandwidth of the HOE, the foci will fall along a line at an angle to the HOE, with different 
wavelengths having focal distances inversely proportional to their wavelength. The spectrum will 
rotate with the HOE, defeating one of the biggest advantages of using the HOE, to simplify the 
scanning design. In order to capture this spectrum while rotating, a line of detectors or focal 
plane optics must rotate at the same rate. Berkoff et. al. reported on an experimental Raman 
lidar design to capture the elastic backscatter and three Raman scattered wavelengths, using the 
HOE as both a telescope and a spectrograph. 

Summary and Conclusions 

We have shown that large HOEs can be used effectively as conical scanning lidar telescopes 
which are mechanically much simpler and more compact than designs using conventional 
telescopes and scan optics. When an HOE is made with a point source object beam diverging and 

24 



normal to the film, and a collimated reference beam at some angle to the film, the object beam 
defines the focal length and the rotation axis for scanning, while the reference beam defines the 
scan cone half-angle. A conical scan pattern is generated when the HOE is rotated in its own 
plane about the optic axis that goes through the focal point. Low energy density laser pulses can 
be transmitted through the HOE, which acts as the final collimating and beam steering optic. 
Two prototype lidar systems, one using a reflection HOE with a 532 nm laser transmitter, and 
one using a transmission HOE with a 1064 nm laser transmitter have been built and successfully 
tested. Angular resolutions as small as 160 ^rad have been obtained, allowing HOEs to be used 
as small FOV receivers. Holographic plane gratings can also be used to conically scan 
conventional telescopes or static HOEs provided care is taken to suppress possible zero and 
negative first order diffracted light if the grating is used to also scan the outgoing laser beam. 
Several years of use have shown the HOE assemblies to be robust and reliable. Future 
developments include scaling to meter apertures and larger, increasing angular resolution, and 
multiplexing HOEs to utilize multiple wavelengths or multiple fields of view so as to negate the 
need to move the HOE in order to scan. 

Acknowledgements 

The authors wish to thank the following programs and offices for their support of this research: 
The NASA Goddard Director's Discretionary Fund, Donald Friedman and Nancy McClennan of 
the NASA SBIR program office, Nona Cheeks, Joseph Famiglietti, and Anel Flores of the 
NASA Goddard Technology Commercialization Office, Lisa Callahan of the NASA Cross- 
Enterprise Program, and Steve Mango of the Integrated Program Office. We thank Barry Coyle 
for helping to design and build the first laser for PHASERS, Steve Palm for developing the first 
PHASERS data acquisition software, Kurt Medine, Wayne Welch, and Jeffrey Freemire for 

25 



PHASERS mechanical design support, Alex Leung for HOE testing, Orbital Sciences 
Corporation and Science and Engineering Services, Inc. for HARLIE engineering, fabrication, 
and HOE testing. Special thanks go to David Miller for his help deploying and operating 
HARLIE and to Sangwoo Lee for help in analyzing the HOE test data. 

References 

1. G. K. Schwemmer and T. D. Wilkerson, "Holographic Optical Elements as Conically 
Scanned Lidar Telescopes," in Optical Remote Sensing of the Atmosphere, Vol. 18 of Tech. 
Digest Series, (Optical Society of America, Wash. D. C, 1991), pp. 310-312. 

2. G. Schwemmer, " Conically Scanned Holographic Lidar Telescope," United States Patent 
#5,255,065, issued Oct. 19, 1993. 

3. G. Schwemmer, T. Wilkerson, and D. Guerra, "Compact scanning lidar systems using 
holographic optics," in Optical Remote Sensing for Industry and Environmental Monitoring, 
Proc. SPIE 3504, 51-59 (1998). 

4. G. Schwemmer, D. Miller, T. Wilkerson, D. Guerra, and R. Rallison, "NASA lidar uses 
HOEs for lightweight scanning," Laser Focus World, 38, No. 6, pp. 141-147, June 2002. 

5. N. Takeuchi, H. Shimizu, Y. Sasan, and N. Sugimoto, "LAMP lidar for wide-area air 
pollution monitoring," in Conf. Abstracts, ICf^ International Laser Radar Conf, Oct. 6-9, 
1980, Silver Spring, MD, p. 16-17. 

6. D. Morancais, R. Sesselmann, G. Benedetti-Michelangeli, and M. Hueber, "The Atmospheric 
Lidar Instrument (ATLED)," Acta Astronautica, 34, 63-67 (1994). 

7. D.N. Whiteman, S.H. Melfi, and R.A. Ferrare, "Raman lidar system for the measurement of 
water vapor and aerosols in the earth's atmosphere," Appl. Opt., 31, 3068-3082 (1992) 



26 



8. E. E. Uthe, N. B. Nielsen, and T. E. Osberg, "Airborne scanning lidar observations of aircraft 
contrails and cirrus clouds during SUCCESS," Geophys. Res. Lett., 25, 1339-1342 (1998). 

9. P. A. Hwang, W. B. Krabill, W. Wright, R. N. Swift, E. J. Walsh, "Airborne scanning lidar 
measurement of ocean waves, Remote Sensing of Environment, 73, 236-246 (2000). 

10. S. P. Palm, S.H. Melfi and D. L. Carter, "New airborne scanning lidar system: applications 
for atmospheric remote sensing, Appl. Opt., 33, 5674-5681 (1994). 

11. J. L. Irish and W. J. Lillycrop, "Scanning laser mapping of the coastal zone: the SHOALS 
system," ISPRS J. of Photogramm. & Remote Sensing, 54, 123-129 (1999). 

12. J. G. Hawley, R. Targ, S. W. Henderson, C. P. Hale, M. H. Kavaya, and D. Moerder, 
"Coherent launch-site atmospheric wind sounder: theory and experiment," Appl. Opt., 32, 
4557-4568 (1993). 

13. M. Bennett, S. Sutton, and D. R. C. Gardiner, "Measurements of wind speed and plume rise 
with a rapid-scanning Lidar," Atmos. Environ., 26A, 1675-1688 (1992). 

14. C. Werner, P. H. Flamant, O. Reitebuch, F. Kopp, J. Streicher, S. Rahm, E. Nagel, M. Klier, 
H. Herrmann, C. Loth, P. Delville, P. Drobinski, B. Romand, C. Boitel, D. Oh, M. Lopez, M. 
Meissonnier, D. Bruneau, A. Dabas, "Wind infrared Doppler lidar instrument," Opt. Eng. 40, 
115-125(2001). 

15. HJHHHi^H-' "The Multi -center Airborne Coherent Atmospheric Wind Sensor," Bull. 
Amer. Meteor. Soc, 79, 580-598 (1998). 

16. E. W. Eloranta and D. K. Forrest, "Volume imaging lidar of the convective structure 
surrounding the flight path measuring aircraft," J. Geophys., Res., 97, 18,383-18,393 (1992). 

17. W. P. Hooper and L. U. Martin, "Scanning lidar measurements of surf-zone aerosol 
generation," Opt. Eng., 38, 250-255 (1999). 



27 



18. W. E. Eichinger, D. I. Cooper, P. R. Forman, J. Griegos, M. A. Osbom, D. Richter, L. L. 
Tellier, and R. Thornton, "The development of a scanning Raman water vapor hdar for 
boundary layer and tropospheric observations," J. of Atmos. Oceanic Technol., 16, 1753- 
1766 (1999). 

19. B. M. Gentry, H. L. Chen, S. X. Li, "Wind measurements with 355-nm molecular Doppler 
lidar," Opt. Lett. 25 (17), 1231-1233 (2000). 

20. Y. D. Chen and J. Ni, "Dynamic calibration and compensation of a 3-D laser radar scanning 
system," IEEE Trans. Robotics Automat., 9, 318-323 (1993). 

21. J. B. Blair, D. L. Rabine, and M. A. Hofton, "The laser vegetation imaging sensor: a 
medium-altitude, digitisation-only, airborne laser altimeter for mapping vegetation and 
topography," ISPRS J. of Photogrammetry and Remote Sensing, 54, 115-122 (1999). 

22. D. H. Close, "Optically recorded holographic optical elements," in Handbook of Optical 
Holography, H. J. Caulfield, Ed., (Academic Press, Inc. 1979), pp.573-585 jjjjjjjjjjjjjjjj^^ 

23. R. D. Rallison, "Holographic optical elements (HOEs) in dichromated gelatin (DCG): 

. progress," in Applications of Holography, L. Huff, Ed. Proc. SPIE 523, 292-295 (1985). 

24. C. Gilbreath, A. E. Clement, and J. W. Wagner, "Holographic Optical Elements which 
diffract efficiently in the near-infrared for operation with laser diodes for spacecraft 
communications," in Practical Holography III, T. H. Jeong, Ed. Proc. SPIE 747, 43-50 
(1987). 

25. D. V. Guerra, A. D. Wooten, Jr., S. S. Chaudhuri, G. K. Schwemmer, and T. D. Wilkerson, 
"Prototype Holographic Atmospheric Scanner for Environmental Remote Sensing ", J. of 
Geophysical Research, 104, No. D18, 22,287-22,292 (1999). 



28 



26. R. D. Rallison and S. R. Schicker, "Wavelength compensation by time reverse ray tracing," 
^^■H^H Proc. SPm, 2404, p. 217 (1995). 

27. R. D. Rallison and S. M. Arnold, "Wavelength compensation at 1.064(i. using hybrid optics," 
in Diffractive and Holographic Optics Technology III, I. Cindrich, S. A. Benton, Eds., Proc. 
SPIE, 2689, 217-226 (1995). 

28. G. K. Schwemmer, "Holographic airborne rotating lidar instrument experiment," 19th 
International Laser Radar Conference, U. Singh, S. Ismail, G. K. Schwemmer, Eds. 
(NASA/CP-1 998-20767 1/PT2, NASA, 1998), pp. 623-626. 

29. T. A. Berkoff, D. N. Whiteman, R. D. Rallison, G. K. Schwemmer, L. Ramos-Izquierdo, and 
H. Plotkin, "Remote detection of Raman scattering by use of a holographic optical element as 
a dispersive telescope", Optics Letters, 25, 1201-1203 (2000). 

30. G. K. Schwemmer, "Shared-aperture multiplexed holographic scanning telescopes," 
Photonics Tech Briefs, 25, No. 7, p. 18a, July 2001 

31. G. K. Schwemmer, "Methods and systems for collecting data from multiple fields of view," 
U. S. Patent No. 6,479,808, awarded Nov. 12, 2002, . 

32. H. Kogelnick, Bell Syst. Tech. J. 48, 2909-2947 (1969). 

33. G. K. Schwemmer and T. Wilkerson, "Development of a holographic telescope for optical 
remote sensing," in Spin-Off Technologies from NASA for Commercial Sensors and Scientific 
Applications, ^^HProc. SPIE, 2270, 40-47 (1994). 

34. T. D. Wilkerson, J.A. Sanders, I.Q. Andrus, G.K. Schwemmer, D.O. Miller, D. Guerra, J. 
Schnick, and S.E. Moody, "The HOLO series: critical ground-based demonstrations of 
holographic scanning lidars", in 2" Asia-Pacific Symposium on Remote Sensing, U. N. 
Singh, T. Itabe, N. Sugimoto, eds., Proc. SPIE, 4153, 63-69 (2001). 



29 



35. G. K. Schwemmer, T. D. Wilkerson, J. A. Sanders, D. V. Guerra, D. O. Miller, S. E. Moody, 
"Ground based operational testing of holographic scanning lidars," in Advances in Laser 
Remote Sensing, A. Dabas, C. Djth and J. Pelon, eds., (Ecole Polytechnique, publisher, 
France, 2001), pp. 69-72. 

36. J. A. Sanders, T. D. Wilkerson, G. K. Schwemmer, D. O. Miller, D. Guerra, S. E. Moody, 
"Comparison of two lidar methods of wind measurement by cloud tracking," presented at the 
20* International Laser Radar Conference, Vichy, France, 10-14 July 2000. 

37. T. D. Wilkerson, I. Q. Andrus, G. K. Schwemmer, D. O. Miller, "Horizontal Wind 
Measurements using the HARLIE Holographic Lidar", in Optical Remote Sensing for 
Industry and Environment Monitoring II, U. N. Singh, ed., Proc. of SPIE, 4484, 64-73 
(2002) . httD://harlie.gsfc.nasa.gov /. 

39. R. D. Rallison and S. R. Schicker, "Polarization properties of gelatin holograms," in | 
Proc. SPIE, 1667, 266-275 (1992). 



30 



Figures 



H.O.E. 




Fig. 1. Exposure geometry (a), and reconstruction geometry (b) for a relection HOE. 



31 



undiffracted 
by Master 



+1 diffraction 
by Master 



V 



Copyfiim 



Master HOE 



Mirror on 
translation stage 



\l 



Laser 



/ \ 

Cyiindrical lens '^ ^ Splierical lens 



Fig. 2. Copying an HOE from a master. The laser beam is formed into a sheet (long dimension is into the 
paper) before scanning the master/copy assembly. 



32 




Fig. 3. Photograph of several HOEs and holographic gratings designed for lidar applications. 



33 



.Ml ,rSii- 




Fig. 4. Geometry for using a transmission 
HOE to scan the transmitted laser as well as 
the receiver FOV. 



34 



a 



X! 
I 



Energ} diffracted by 
532 nin HOE (relative) 




490 500 



560 570 



Wavelength (nm) 
Fig. 5. Relative spectral response of HOE #1R. The curve is a spline fit to the data points. 



35 



OfF-Axis Parabola 



Location of lens and meter for 
diffracted power measurement 



CCD Camera 




Fig. 6. Setup for measuring diffractiisn efficiency and spot size. 



36 



o.ao 



n a o o a a a n 

0.70 ° a a a a ° " ° ° a 



^ 0.60 

ti 

M 

S 0.50 



0.40 
O.JO 



35 40 45 

lncid«nt on^l« (deg) 



0.80 
0.70 



0.60 ■£• 

M 

0.50 ;: 



o.+o 



0.30 



50 



Fig. 7. Diffraction efficiency (squares) and 
spot size (triangles) as a function of incident 
angle for HOE#7R. 



37 



Flat Mirror 



OfF-Axis Parabola 




Theodolite 
order light (2 measurement positions) 



Fig. 8. Grating diffraction angle test setup. 



38 



COLOK TABLE 

16 Leuels 
23S-2Sl-» 
Z20-234-t 
284-219-* 
188-ZB3-> 
173-187-» 
157-172-* 
141-1S6-* 
126-140-» 
il8-125-> 
94-1B9-' 

78-93-» 

63-?7-»' 

47-62^ 

31-46-> 

16-3B-* 
8-lS-> 
Contouring Node = ftutopiatic Scaling to Peak 

Fig. 9. CCD camera image of focal spot of one of 
the preliminary copies of HOE #2. Tlie reticle 
circle diameter is 270 um. 



(% 




^ 





39 



tn 

c 
o 

la. 

o 

1 

as 
-i 
O 

E 
o 
u. 
o 

UJ 

t- 

uu 

< 



3000 

2700 

2400 

2100 

1800 

1500 

1200 

900 

600 

300 





2.25 



2.00 



<M 

0) 



1 


1.75 


E 




1.50 


m 
oc 

3 




1.25 


CC 
Ul 

n 




1.00 


< 




0.75 


< 

3 




0.50 


C3 

Z 
< 



10 20 30 40 50 60 70 80 
FRACTION OF TOTAL ENERGY (%) 

Fig. 10. Encircled energy function of focal spot for HOE #1R. 



I 0.25 



90 100 



0.00 



40 




25% 



35% 45% 55% 65% 75% 
Fraction of Total Energy (%) 



85% 95% 



Fig. 11. Encircled energy function for HOE #2. 



41 




Fig. 12. Early version of tlie first atmospheric 
lidar using a scanning reflection HOE. 



42 




PHASERS ASSEMBLY 



Fig. 13. Mechanical drawing of tlie new PHASERS transceiver. 



43 




Fig. 14. HARLIE system transceiver (left) and 
data system (right). 



?^- 



44 



ex. 



«*0 — ' 



AIRCRAFT 
SKIN 



usat 

>..''.:...".:.. <. PLOOR 




Fig. 15. Position of HARLIE in an aircraft, 
suspended from the floor. 



45 




Fig. 16. A scan of ground-based HARLIE backscatter data taken the night of 9 June 1999, during 
the HOLO-2 campaign (Wilkerson et.al., 2001). 



46 



1 ""-order 

transmitter 

beam 



1 •'-order 
backscatter 



O-order 

V transm^er 

beam ! 




Reflective telescope 



1 "-order 

transmitter 

beam 




O-order 
V — transmiher 
beam 



^^ — Transmission HOE 



O-order 
~ backscatter 



ib) 

Fig. 17. a) Crosstalk from the O-order beam is focused in a conventional telescope used with a 
grating scanner, b) Crosstalk from the O-order beam is not focused in an HOE telescope. 



47 




Figure 18. Two methods of wavelength multiplexing HOEs to maintain static detector 
assembles, a) The foci are all placed on the rotation axis, b) A rotating plane grating scanner 
followed by a static multiplexed HOE. 



48 



355 nm light 




Renedlon HOE 



Transmission i-iOE 



1064 nm light 



Fig. 19. Wavelength multiplexed design 
concept using a 355 nm reflection HOE on top 
of a 1064 nm reflection HOE. 



49 



Figure Captions 

Fig. 1. Exposure geometry (a), and reconstruction geometry (b) for a relection HOE. 

Fig. 2. Copying an HOE from a master. Tlie laser beam is formed into a slieet (long dimension is into the 
paper) before scanning the master/copy assembly. 

Fig. 3. Photograph of several HOEs and holographic gratings designed for lidar applications. 

Fig. 4. Geometry for using a transmission HOE to scan the transmitted laser as well as the receiver FOV. 

Fig. 5. Relative spectral response of HOE #1R. The curve is a spline fit to the data points. 

Fig. 6. Setup for measuring diffraction efficiency and spot size. 

Fig. 7. Diffraction efficiency (squares) and spot size (triangles) as a function of incident angle for HOE#7R. 

Fig. 8. Grating diffraction angle test setup. 

Fig. 9. CCD camera image of focal spot of one of the preliminary copies of HOE #2. The reticle circle 
diameter is 270 um. 

Fig. 10. Encircled energy function of focal spot for HOE #1R. 

Fig. 11. Encircled energy function for HOE #2. 

Fig. 12. Early version of the first atmospheric lidar using a scanning reflection HOE. 

Fig. 13. Mechanical drawing of the new PHASERS transceiver. 

Fig. 14. HARLIE system transceiver (left) and data system (right). 

Fig. 15. Position of HARLIE in an aircraft, suspended from the floor. 

Fig. 16. A scan of ground-based HARLIE backscatter data taken the night of 9 June 1999, during the 
HOLO-2 campaign (Wilkerson etal., 2001). 

Fig. 17. a) Crosstalk from the 0-order beam is focused in a conventional telescope used with a grating 
scanner, b) Crosstalk from the 0-order beam is not focused in an HOE telescope. 

Fig. 18. Two methods of wavelength multiplexing HOEs to maintain static detector assembles, a) The foci are 
all placed on the rotation axis, b) A rotating plane grating scanner followed by a static multiplexed HOE. 

Fig. 19. Wavelength multiplexed design concept using a 355 nm reflection HOE on top of a 1064 nm reflection 
HOE. 



50 



Table 1 



Table 1 . Partial List of the HOEs and gratings manufactured and tested in this program. 

Item #^ Wavelength (nm) Diameter (mm) Focal Length (mm) Angle 1 Angle 2* 



IR 


532 


404 


1300 


2 


1064 


404 


1016 


3 


1064 


404 


00 


4 


770 


404 


00 


5 


532 


404 


00 


5R 


532 


404 


1295 


6a 


1047 


202 


00 


6b 


1047 


202 


760 


7a 


1047 


254 


00 


7b 


1047 


254 


760 



45 





45 





45 





45 





36 





43 





22.5 








22.5 


22.5 








22.5 



^ An R following the number indicates a reflection hologram, otherwise it is a transmission hologram. 

* For HOEs, angle 1 refers to the coUimated beam axis and angle 2 refers to the focusing axis, relative a normal 

to the substrate. 



51 



Table 2 



Table 2. Test measurement results for holographic optics of table 1. 



Itein# 



Focal 
lengt 
h±2 
(mm) 



Spot 
size 

i\xrad 

)± 

5% 



Spot 

size 

FWHM 

(fxrad) 

±5% 



Zero 

order 

efficiency 

±0.2(%) 



1*" order 
efficiency 

±2(%) 



Diffraction 
angle ±0.1 
(degrees) 

for 

smallest 

focal spot 



Comments 

Errors are 

given column 

in headings 

unless 
otherwise 

noted. 



IR 

2 
3 
4 
5 
6R 



1295 1200 



1016 

N/A 
N/A 
N/A 
1306 



1306 



180 

N/A 
N/A 
N/A 
630 



486 



N/A 
N/A 
N/A 
350 



269 



5 
9.5 

7 
13 

~3 



7a 


N/A 


N/A 


N/A 


n 


N/A 


N/A 


N/A 


@904nm 








7b 


756.2 


225 X 


56 X 






262 


138 


8a 


N/A 


N/A 


N/A 


8b 


760.1 


135 X 


43 X 






285 


149 



1.3 
11 

7.4 

2.9 
6.4 



59 

86 

73 
84 
74 
73 



72 



91 

76 

84 

90 
86.5 



42.0±0.5 

45.0 
45.0 
45.0 
36.4 
40.1 



40.1 



PHASERS 
HOE 
HARLIE HOE 



-7% specular 
reflections from 
internal 
surfaces. 
The best focus 
occurred at this 
angle. 



22.8 
20.3 

23.2 

22.1 
22.7 



* If two numbers are given they describe the major and minor axis of an ellipse, if one number it describes a 
circle. Either conic contains 86.5% of the energy diffracted into the first order. 



52