Whitepaper

Excelitas Technologies White Paper on

High Energy Laser Optics

Jack Hickman Trey Turner Roger Kirschner

Business Development Manager VP Research and Development Business Development Manager
Excelitas Technologies Corp. Excelitas Technologies Corp. Excelitas Technologies Corp.

April 23rd, 2021

1.0 Overview
High Energy Laser (HEL) systems have become a reality in the military domain. As this technology continues to increase in
Power output, the Optical Manufacturing industry will be pressured to produce optics with levels of performance and reliability
that are not commonplace in the industry today. The system level characteristics of Beam Quality, Throughput, Reliability,
SWaP and Cost will continue to be at the forefront of the discussions for the current and future generations of HEL weapons.

This paper walks through how the Specifications and Manufacturing Methods of the Optical Components in these HEL systems
affect each of the system level characteristics. The purpose of this paper is to clarify to the Designer, Supply Chain Managers,
Program Managers, and Material Buyers the options and tradeoffs available to them from an Optical perspective. The goal is
to save time and money from a system design perspective through a more efficient understanding of what is possible.

Enabling the future through light.

©2021 Excelitas Technologies Corp. All rights reserved. The Excelitas logo and design are registered trademarks of Excelitas Technologies Corp. All other trademarks not owned by Excelitas Technologies or its subsidiaries
that are depicted herein are the property of their respective owners. Excelitas reserves the right to change this document at any time without notice and disclaims liability for editorial, pictorial or typographical errors.

Page 1 of 18

www.excelitas.com
1 Introduction
High Energy Laser (HEL) systems have been in development for many decades, but recently are gaining traction as a viable
military tool. As modern weapons become more powerful, transportable, and effective, it is important first to understand
how recent history has shaped the industry today.

In the early 2000s, the National Ignition Facility (NIF) program at the Lawrence Livermore National Labs demonstrated
the extreme power that chemical lasers can generate, initiating the modern era of high-powered lasers. This system
combines 192 lasers to achieve up to 500 Terawatts of peak power, enough to induce nuclear fusion reactions – the goal
of the program. To this day, the NIF Laser remains the largest and highest-energy laser in the world [1].

In 2009, Directed Energy Systems and Northrop Grumman demonstrated a more transportable approach by combining
seven 15kW electric lasers for the Joint High Power Solid-State Laser (JHPSSL) that achieved weapons-grade power of
105kW with very good efficiency and beam quality [2]. This program was key in proving that military-grade transportable
systems were in fact possible, and sparked the industry to continue progressing the technology.

Entering the 2020s, Defense Contractors are focusing on demonstrating ever more powerful laser weapon systems.
Examples of the intended function of these systems are as follows: to defend military bases or battleships from drone
attacks, defend a high-profile plane from attack, shoot down ICMB and hypersonic ballistic munitions, protect satellites
in orbit, and even for use in air-to-air combat. These applications and others will necessitate an extremely accurate and
reliable laser system with increasing levels of power requirements. The key system level characteristics to consider as the
technology moves forward are Beam Quality, Throughput, Reliability, SWaP and Cost.

The overall system can benefit from a better understanding of the optical components’ impact from a manufacturing
perspective, as controlling the components is the key to unlocking the best system performance possible. To achieve this,
system level integrators will need work closely with their optics suppliers to understand best how to specify their
components and sub-assemblies.

Excelitas Technologies Corp. has been involved with sub-components for high-energy lasers for decades, having
manufactured HeNe lasers and optics for laser systems since 1993 as Research Electro-Optics, Inc. (REO) in Boulder,
Colorado. REO’s notable history includes commercializing the use of Ion Beam Sputtering (IBS) coating technologies for
the most prolific Ring Laser Gyroscope systems in the world, as well as for extremely high-power industrial lasers. REO was
a key optics supplier for both the NIF and JHPSSL programs, and continues to be involved with many next-generation high-
energy laser systems. Recent testing on Excelitas HEL optics showed outstanding performance during a 150kw system test.

Excelitas Technologies Corp. acquired REO in 2018. Excelitas continues to operate at the forefront of optics manufacturing
technology, and maintains the most robust understanding of Laser Induced Damage Threshold in the industry. This is the
reason why Excelitas’ optics continue to be the most reliable for high-energy laser systems, exemplified through continuous
use in the NIF laser almost 15 years after inception.

©2021 Excelitas Technologies Corp. All rights reserved. The Excelitas logo and design are registered trademarks of Excelitas Technologies Corp. All other trademarks not owned by Excelitas Technologies or its subsidiaries
that are depicted herein are the property of their respective owners. Excelitas reserves the right to change this document at any time without notice and disclaims liability for editorial, pictorial or typographical errors.

Page 2 of 18

www.excelitas.com
2 System Characteristics
The critical Laser System Characteristics of High-Energy Laser systems are Beam Quality, Throughput, Reliability, and
SWaP-C. These four characteristics play a large role in the overall effectiveness of the laser system performance.

2.1 Beam Quality maintenance intervals needed, overall cost of use, and the
Beam quality refers to how “sharp” the laser spot is when environments in which it is effective. Optics tend not to be
the laser hits a target. Technically, it is how close the spatial the cause for field failures, as optical failures are rarely
distribution of the beam is to an ideal (usually Gaussian) latent failures. However, an understanding of an optic’s
beam. Better beam quality results in more energy in the stability can quantitatively be understood through
beam and more power transmitted to the target. This measurements of Surface Quality inspection, Temperature
impacts the performance characteristics of effective range, and Humidity Sensitivity Testing, and Durability Testing.
spot size, quality of the effect, and speed of the effect.
There are factors such as distance to target, atmospheric 2.4 SWaP-C (Size, Weight, Power, Cost)
turbulence and thermal blooming [3] that play into effect As with any military system, Size, Weight, Power and Cost
that can be out of one’s direct control. Beam Quality can are important design considerations. With High Energy
be influenced by tightly controlling the optical Laser systems, there is a general understanding that larger
specifications Surface Figure and Absorption. systems allow for lasers that are more powerful. Thinking
of the NIF Laser system, this logic holds. The problem,
2.2 Throughput however, is that these HEL systems must be mobile enough
The next critical aspect of high-energy laser systems is the for battlefield situations. Depending on the application, it
Throughput of the system. Throughput is a measure of how can be more (vehicle mounted) or less mobile (base
much power the system can handle. A strong power source defense), but the larger the system becomes, the more
is meaningless if only a fraction of the power is transmitted difficult to move and more power is needed to operate.
through the system, so high Throughput yields higher Recent advances in fiber laser technology have allowed
power transmitted to the target. Throughput impacts much smaller packages to produce much higher power
overall performance characteristics of effective range, lasers. This has helped reduce the overall size of many laser
quality of the effect, speed of the effect, and size of the systems, but there is still a drive to have systems with exit
system. The optical specifications of Aperture Size, apertures greater than 550mm. As these systems grow, so
Reflectivity and Transmission, and Absorption play a large must all of the optics in the system in order to
role in maximizing throughput. Laser Induced Damage accommodate the larger beam size. With optics, the larger
Threshold must also be considered; optics will fail as the parts get the more difficult and expensive it becomes
Throughput increases, unless the optics are designed to to manufacture.
withstand the energy.
The following section describes the optical specifications
that influence these system level characteristics. The
2.3 Reliability
manufacturing methods to achieve these optical
Potentially the single most important characteristic of an
specifications will be explored in a later section. The final
HEL System is its Reliability. If a system fails in the field, it
section describes tradeoffs when prioritizing each system
can put many lives in danger. The length of time a system
characteristic.
is expected to operate under intended use conditions is the
main concern of many end users. Reliability influences the

Page 3 of 18

www.excelitas.com
3 Optical Component Specifications
Optical Specifications heavily influence the laser system characteristics. The optical specifications to consider are Laser
Damage, Surface Quality, Absorption, Spectral Performance, Stability & Durability, Aperture Size and Surface Figure. Each
specification can influence one or more system characteristic, so it is necessary to understand the interplay between the
specifications.

3.1 Laser Damage

Laser Damage, or Laser Induced Damage Threshold (LiDT), is the level of laser fluence that
an optic can safely withstand without damage. Lasers are either Pulsed or Continuous
Wave (CW), and the failure mechanisms are different between the two types. An example
of a damage event is shown in Figure 1.

For Pulsed lasers, the damage mechanisms of nonlinear absorption and multi-photon
absorption are key. This has to do with the material’s ability to pass electrons through the
bulk material. If too many electrons are absorbed, a damage event occurs. The material’s
band gap, which strongly relates to the refractive index, is the main driver in the ability
to pass electrons. The substrate and optical coating materials must be considered, which
plays a heavy role in how thin film coatings are designed for these applications. Figure 2
shows example bandgap values for various optical materials and the associated LDT.

Material Bandgap (eV) n LDT (J/cm2)

Ge 0.7 4.4+0.15i 0.24
λ = 1030 nm (1.2 eV)
ZnSe 2.5 2.49 0.53 τ = 500 fs
Ta2O5 4.1 2.1 2.2
Gallais and Commandré
HfO2 5.3 1.9 3.4 Appl. Opt., 53, A186 (2014)
Al2O3 6.5 1.65 4.2
SiO2 7.5 1.46 6.3
Figure 2. Bandgap of various material types.

For Continuous Wave Lasers, the dominant damage mechanism is through linear
absorption, so thermal effects are key. The thermal conductivity of the bulk and thin film
materials must be considered, and again influences what materials are chosen for a thin Figure 1. Laser Damage Example
film coating.

However, even when the correct materials are chosen and the thin film design is optimized, laser damage events can still
occur. The damage mechanism is the enhancement of E-field by the curvature of defect-initiated nodules that lowers the
damage thresholds to below intrinsic values. Simply put, if the beam interacts with a defect, damage is more likely than
on a defect-free spot on the surface. The sparse nature of defects makes the evaluation of laser damage over a surface a
statistical challenge [4].

Page 4 of 18

www.excelitas.com
3.2 Surface Quality
Surface Quality is the optical specification that
quantifies the level of defects on an optical surface. The
“better” the surface quality, the fewer defects are
present.

There is work being done to better understand how

defects correlate to laser damage. It must be
understood that there is never a zero chance of damage
at any level of laser fluence, regardless of defect size
[4]. Figure 3 shows an example of the probability of
Damage events at various fluence levels.

In a separate Excelitas study, it was shown that a 10µm

defect will initiate damage with approximately 50%
probability at 50J/cm2 in a typical HfO2/SiO2 High
Reflector, illuminated with a 20ns pulse at 1064nm [5].
Figure 4 shows an example of probability of damage
from high fluence over a surface. Note the profile tends Figure 3. Probability of Damage Events at various fluence levels [1]
to follow the Gaussian shape, as would be expected.

To give the best chance of not having a damage event occur on

an optic, the optimal solution is to have zero defects on the
optical surface. This is next to impossible to achieve in realistic
manufacturing practices, but the goal is to minimize the amount
of particles that are exposed to the component. Generally, as
surface quality gets better, Laser Damage and Durability improve,
but cost goes up.

Excelitas can offer the standard LiDT performance at various

wavelengths shown below in Figure 5, depending on aperture
size and material. Note that more robust specifications can be
achieved, but the values shown are an expectation of baseline
Figure 4. Damage location probability chart for
performance.
high fluence laser optics
Pulsed - 20ns CW
λ
HR AR HR AR
1064nm 50 J/cm 2
30 J/cm 2
20 MW/cm 2
10 MW/cm2
532nm 30 J/cm2 15 J/cm2 - -
355nm 10 J/cm2 5 J/cm2 - -
Figure 5. Common LiDT Performance at various wavelengths.

Page 5 of 18

www.excelitas.com
3.3 Absorption
An optic in a laser system can either reflect, transmit or Method Accuracy Sensitivity
absorb the light as it interacts with the optic. The goal with Photothermal CPI 20% 0.1ppm
optics in high-energy laser systems is to absorb as little of Calorimetry 1% 10ppm
that light as possible, particularly in Continuous Wave
Spectrophotometry 1% 1000ppm
systems. Light gets absorbed either via the bulk optical
Figure 7. Absorption measurement methods.
material, through defects on the surface of the optic that
cause the light to scatter, or by the materials that make up When absorption is optimized, Continuous Wave Laser
the optical coating [4, 6]. Minimum achievable absorption Damage performance improves. Interestingly, Pulsed Laser
values at various wavelengths are shown below in Figure 6 Damage performance decreases, due to the materials used
for both High Reflector and Anti-Reflective coatings. in the optical coating. It is very important to specify which
type of system the optic will be used in, as the
λ HR AR manufacturing process and materials are not the same
1064 nm 1 ppm 3 ppm depending on the use. Additionally, multi-spectral
940 nm 2 ppm 3 ppm performance will decrease as absorption is optimized. This
532 nm 10 ppm 20 ppm will be discussed more in the following section.
355 nm 0.05% 0.07%
Figure 6. Minimum achievable absorption levels.

One of the main differentiating technologies in use to

minimize absorption is IBS Coating technology. IBS
Coatings can produce absorption levels down to 1ppm
repeatedly, whereas an E-Beam coating would tend to be
more in the 100ppm levels.

When working with absorption levels in the 1ppm level,

non-standard measurement methods are required. The
most sensitive method is called Photothermal Common
Path Interferometry (CPI), which is sensitive down to
0.1ppm. While the accuracy of this system is ~20%, it is the
best way to confirm absorption levels necessary for the
best High Energy Laser optics. Other measurement
methods’ performance is shown in Figure 7.

3.4 Spectral Performance

Reflection and transmission levels are extraordinarily important in High Energy Laser systems. As each optic will not reflect
100% of the incoming beam, there is a leaking effect where the beam power exiting the system could end up being
dramatically lower than what came out of the laser source. This means lower power on target, and a less effective system.

For a single optic at a single wavelength, it is relatively easy to achieve a high reflector with up to 99.9995% reflectivity.

Page 6 of 18

www.excelitas.com
Take the example shown in Figure 8, which is a 33-layer
Ta2O5/SiO2 design. It has an average reflection of
~99.997% across the design region centered near
1064nm, Absorption of <2ppm, Design Thickness of
~5µm (resulting in a low stress coating, short run time,
and cheaper coating), and a Surface Quality much
better than 10-5. The resulting optic would have very
good laser damage resistance and would perform
extremely well for the primary band. This would be a
great solution if the optic only needed to reflect a
single band.

The issue is that most High Energy optics are designed

to do more than just reflect the primary waveband;
usually there are 3-4 different wavebands. This type of
design requires compromises to reflectivity levels,
absorption, and laser damage in order to be produced.
Figure 8. R% (above) and design (below) of a Single Band High reflector

Take the example shown in Figure 9 of a multi-spectral

high reflector, which is a 200-layer Ta2O5/SiO2 design.
This has an average reflectance of 99.997% at the
design wavelength near 1064nm, Absorption of
~20ppm, design thickness of 21µm (high stress, long
run, expensive), and surface Quality around 20-10.

In order to optimize multi-spectral performance of this

coating, the reflection levels of the other bands of
interest must decrease, the overall laser damage
resistance will be lower, and the surface figure control
would decrease. This would be a very expensive optic,
and generally will be very hard to produce repeatedly.

When specifying multispectral designs, it helps to

prioritize bands clearly so if tradeoffs must be made
during the thin film design, the manufacturer can do so
appropriately. In addition, it helps to not over specify
Figure 9. R% (above) and design (below) for multi-spectral high reflector.
R% in secondary bands, as it will be very difficult to
achieve high levels of performance across each band.
Lastly, if transmission bands can be used in place of
additional reflection bands, it tends to be easier to
manufacture for multispectral coatings.

Page 7 of 18

www.excelitas.com
The alternate way to achieve performance with multiple Since there are multiple optical components in High Energy
wavelengths of interest is to create secondary systems that Laser systems, it is extremely important that all optics are
are boresight aligned to the primary system. This allows the transmitting or reflecting as much as possible on an
primary optics to be optimized for best performance, while individual basis for a given wavelength, as even one low-
the secondary system gives up some performance in order performing optic can reduce the entire system
to accommodate more functions. If a design tries to include performance. Take the below chart, Figure 10, as a trivial
too many functions in a single optical train, the coatings example being considered at one wavelength for an eight
become almost impossible to manufacture, especially at optic system in reflection.
volume.

Number of Optics 8 8 8 8 8 8 8 8
% Reflection Avg. 95.000% 96.000% 97.000% 98.000% 99.000% 99.900% 99.990% 99.999%
System %R 66.342% 72.139% 78.374% 85.076% 92.274% 99.203% 99.920% 99.992%
Figure 10. Component Transmission quickly affects system Transmission

Notice how moving from an eight optic Average %Reflection of 95.0% to 99.0% raises the overall system %Reflection by
almost 26 percentage points. From a different perspective, the difference between an eight optic average Reflection of
99.9% to 99.999% is only a 0.789 percentage point increase in system performance.

The key here is that after a certain point, there are diminishing returns when trying to squeeze out slightly better reflection
on each optic. If system designers work with Excelitas during the design phase, we can help to suggest specifications for
each optic that will allow for the optics to overall be cheaper, better performing, and more easily made in volume.

3.5 Stability and Durability

The stability and durability of an optic refers to the degree transmission performance is measured before and after
of exposure to abrasion, humidity, thermal shocks, etc. humidity exposure, any spectral shift change is the
that an element can withstand without damage. The resulting metric.
factors that play into this stability are coating materials,
raw material CTE, and coating process. This level of stability “Pin holes” or process defects in a coating can cause small
and durability can be quantified through various spots in the coating to be more susceptible to humidity
measurements and tests. absorption, leading to reduced reliability. Through Surface
Quality inspection, this reliability can be quantified. E-
For stability, the goal is that an optic can be exposed to Beam coatings are more susceptible to this defect, as small
various environmental conditions without impacting pin-hole defects are inherent to this coating type. IBS
optical performance. The main environmental factors to coatings are more durable because they do not have these
consider are temperature and humidity changes. defects.

For humidity, the first metric that quantifies stability is Concerning temperature stability, most coating materials
known as the “approximate humidity shift”, also known as are resistant to thermal changes up to the phase change
“Wet/Dry shift”. When optics are exposed to high temperatures. E-Beam coatings are generally applied at
humidity, there can be a shift in optical performance due temperatures well above any normal system operating
the chemical sensitivity of the materials. The reflection and temperature, so once the optics have cooled, they are

Page 8 of 18

www.excelitas.com
quite stable. IBS Coatings are applied at room temperature, this is not a common failure, as the coatings all impact a
but they are still very resistant to temperature changes. Compressive stress on the optic. The stress is quantified
through surface figure measurements, so an optical
Any temperature sensitivity an optic may experience is coating’s relative stability can be readily understood.
generally due to differential thermal expansion coefficients
between the substrate and the coating. If the stress
relationship changes between the coating and the
substrate, small cracks can form in the coating. This
phenomenon is known as crazing, an example shown to the
right in Figure 11. Crazing is more common in coatings with
Tensile stress, which is possible with E-Beam coatings.
These cracks appear when the differential stress exceeds Figure 11. Example of crazing in a coating.
the tensile yield strength of the coating. For IBS coatings,

3.6 Aperture Size

The aperture size refers to the size of the element and the fraction of its area that is filled by the beam. The larger the
aperture, the better the beam quality and the higher the throughput. Large apertures are required for some of the very
highest power HEL systems, but this larger size usually comes with much higher cost and weight.

With Optics, the cost increases proportionately with the surface area of the part; larger means more expensive. This is
because the material itself is more expensive, the fabrication is more difficult, and the optical coating chamber capacity
is reduced, among other factors. Large optics can also require special equipment to manufacture and usually require
custom tooling to hold or move the optics without damaging them.

Obtaining optical grade material properties across a very large material

boule is difficult for many materials, so the cost will inherently go up.
Additionally, there are IR materials that cannot be grown above specific
sizes, so it is best to consult with your optics supplier about what is
possible early in the design and what material you intend to use. Many
large optics tend to employ lightweighting features to make them easier
©
Schott
to handle without losing rigidity properties, but these bring myriad other
issues with controlling surface figure as well as adding additional cost.
Figure 12. Example of large, lightweighted optic.

For fabrication, it is harder to maintain good surface figure and surface quality across a large optical surface. The
machinery required to manufacture can have size restrictions, so custom machinery is required for large optics. A factor
to consider is that just because an optics supplier is good at small components, it does not mean they will be good at
making large components. It is best to consult with the fabrication shop about what sizes they commonly manufacture.

For coating, there is a similar size restriction with machinery. However, assuming the optic can fit in a coating chamber,
the costs still scale with the area of the part. The larger the optic, the smaller the batch size. The smaller the batch size,
the higher the cost. Large optics also require special tooling to lift and place the optic into a coating chamber as well as
off-set weights that may be necessary to balance the crucible inside the coating chamber that holds the parts.

Page 9 of 18

www.excelitas.com
3.7 Surface Figure
Surface Figure refers to how accurately an element matches its nominal surface shape. Surface Figure affects Beam
Quality and Throughput from a system level perspective. Surface Figure plays a role in both transmissive and reflective
optics, and is one of the largest cost drivers in optical manufacturing.

Surface Figure can be controlled very tightly, both in fabrication and coating processes. However, the tighter the Surface
Figure requirement, the more difficult it will be to make and more expensive the optic will be. There are myriad factors
that impact Surface Figure, but the biggest cost driving factors are substrate material type, size, coating stress, and
lightweighting features. Depending on the combination of factors, the Surface Figure specification may be extremely
difficult to achieve. There are various fabrication methods to control Surface Figure, which will be discussed in detail in
later sections. In coat, there are techniques such as back-side stress compensation coatings, stress relief with high
temperature baking, or having a “pre-figured” surface that becomes “flat” when the coating is applied.

The images below in Figure 13 (left to right) show pre-coat, post-coat, and post-coat-bake surface figure maps of the same
optic. These images highlight the difference between Power, PV, and RMS specifications. Note a Negative Power value
indicates a “Convex” shape in this measurement. These pictures also highlight that the Surface Figure changes
dramatically throughput the manufacturing process. Understanding how to control this change is key to being able to
produce optics with good Surface Figure.

Figure 13. Surface Figure measurements of pre-coat (left), post-coat (middle) and post-coat bake (right).

It should be understood that just because all surfaces in a system are “flat”, it does not mean the beam quality and
throughput will necessarily be good, because other factors come into play. The main take away should be that as aperture
size increases or coating complexity increases, it becomes significantly more difficult to control Surface Figure.

4 Materials and Optical Manufacturing Methods

The Materials chosen and Optical Manufacturing Methods employed determine the level of performance of each Optical
Specification. Each Method affects more than one optical specification, so an entire understanding of the processes from
raw material to final optical coating is required in order to produce a reliable and repeatable result.

4.1 Substrate Material

There is no one single material that is “best suited” for High Energy Laser applications. Each material has its tradeoffs and
best use cases. Depending on the spectral performance needed, many materials are not suitable right from the start,
while others may not be able to take advantage of certain Coating technologies. The four most commonly used substrate
materials in High Energy Laser systems are Fused Silica, Silicon, Silicon Carbide, and Aluminum.

Page 10 of 18

www.excelitas.com
Fused Silica (Figure 14) is an optically pure and transparent material with excellent
properties for polishing. It is good for transmissive components and smaller
mirrors. It is a very common material and most optical manufacturers can machine
it. It is machined using traditional polishing techniques. It can produce the lowest
surface roughness and has outstanding thermal expansion properties.
©
Heraeus
Figure 14. Example of a Fused Silica optic.

Silicon (Figure 15) has excellent thermal and mechanical properties and is suitable for
high quality polishing. It can be diamond turned or machined using traditional optical
manufacturing methods. It has very good thermal conductivity and expansion
properties, and as such is a great high-fluence mirror.
Figure 15. Example of a Silicon optic.

Silicon Carbide (Figure 16) has outstanding thermal and mechanical properties for
lightweighted optics. However, it is extremely challenging to polish due to its Hardness and
only a few companies can manufacture it. It gets expensive very quickly, but it is great for high
©
performance mirrors that need complex lightweighting. Aperture Optical Sciences

Figure 16. Example of a Silicon Carbide Optic.

Aluminum (Figure 17) is easy to machine and has excellent thermal conductivity,
but has poor properties for optical polishing. It is cheap and conducts heat well. It
is generally used as a mirror that does not need outstanding optical capabilities.

Figure 17. Example of an Aluminum optic.

Typical property values of each material type is shown below in Figure 18. Note that there is not one “ideal” material for
High Energy Laser systems, each optic serves a specific use and should be designed accordingly. Silicon, however, should
be considered a very good “middle of the road” material that is suitable for many High Energy Laser applications.

Typical Values
Property
Fused Silica Silicon Silicon Carbide Aluminum
Specific Modulus 25 MPa-m /kg 3
60 MPa-m /kg3
140 MPa-m /kg3
25 MPa-m3/kg

Hardness 522 kg/m3 1100 kg/m3 2800 kg/m3 120 kg/m3

Roughness <0.5A rms 1A rms 3A rms ~20A rms

CTE 1ppm/°C 2.6 ppm/°C 4 ppm/°C 23 ppm/°C

Thermal Conductivity 1.4 W/m-K 148 W/m-K 300 W/m-K 215 W/m-K

Cost $$ $$ $$$ $
Figure 18. Typical Substrate Material Properties for commonly used HEL material.

Page 11 of 18

www.excelitas.com
4.2 Polishing and Finishing Methods
There are various Polishing methods available to achieve desired specifications, depending on the requirements, shape,
and form factor of each optic. Each of these processes is utilized for a specific reason, so the more information that can
be given to the optical supplier during the quote phase, the more custom-tailored the manufacturing process can get.
The outcome of the polishing process influences Surface Figure, Surface Quality, and Surface Roughness; all specifications
that highly influence HEL optical performance, as discussed.

Conventional Pad Polishing (Figure 20) is a full-aperture polishing process

using loose abrasive slurry and polyurethane pads in a spindle, lap, or CNC
machine configuration. It is best suited for medium-quality spherical or
aspheric components. It can produce excellent surface quality and is
relatively fast to complete.

Figure 20. Example of Conventional Pad Polishing.

Pitch Polishing (Figure 19) is a full-aperture polishing using loose abrasive

slurry and formed natural or synthetic pitch, in a spindle or lap
configuration. It is best suited for very high fluence components. It is a
slow process, but it produces extremely low surface roughness and
excellent surface quality. An extension of this is known as “Super Polish”,
that produces <0.5Angstrom rms surface roughness.
Figure 19. Example of Pitch Polishing.

Robotic Polishing (Figure 21) is sub-aperture polishing using a small pad

tool controlled for precise, laterally deterministic material removal. It is
not a full aperture polish, but allows for deterministic and repeatable
Surface Figure control. It is best for freeform or off-axis components. It is
a decent way to make lenses, but cannot compete with a Pitch Polish
surface finish. It leaves a trace in mid-spatial surface figure frequencies.
Zeeko

Figure 21. Example of Robotic Polishing.

Magnetorheological Finishing Polishing, commonly known as MRF, is a

sub-aperture finishing process using magnetically shaped loose abrasive
slurry for precise and laterally deterministic material removal (Figure 22).
MRF is best suited for figure-critical components and is used to correct
Wavefront errors. It is not a standalone polishing process, and is best used
in conjunction with Pitch or Robotic Polishing. It is a very fast process and QED

leaves a great surface quality with decent surface roughness.

Figure 22. Example of MRF Polishing.

Page 12 of 18

www.excelitas.com
Typical property values are shown below in Figure 23 for each polishing process.

Typical Values
Property
Conventional Pad Pitch Robotic MRF
Figure Flexibility Flat, spherical, on-axis asph Flat, spherical Arbitrary Arbitrary

Efficiency/Cost Fast (~ 1hr) Slow (~ 6-8 hrs) Varies Fast (<1hr)

Mid-Spatial Freq Essentially Zero Essentially Zero Some tool print-through Some tool print-through

Surface Quality “0-0” (5/ 2 x 0.016) “0-0” (5/ 2 x 0.016) 10-5 10-5

Roughness ~ 3A rms < 0.5A rms ~5A rms 2A rms

Subsurface Damage ~ 10um - ~10um -

Figure 23. Polishing methods typical values.

The main take away from this is that depending on the specifications on the drawing, the manufacturing shop will choose
the polishing method to match. If a specification that drives that decision (Surface Roughness, Surface Figure, or Surface
Quality) is over specified on a drawing, it could lead to unnecessary cost and lead time added to the optics.

4.3 Thin Film Coating Methods

A high quality Optical coating is a critical aspect in a High Energy Laser system as it is the conduit for reflecting or
transmitting the energy through the system at the correct wavelengths. The optical characteristics that are most
important in an HEL specific coating are absorption, scatter, repeatability, abrasion resistance and the ability to validate
the coating’s performance. There are four main technologies used to produce Optical Coatings: E-Beam/Thermal
Evaporation, Plasma/Ion-Assisted Deposition, DC/RF Magnetron Sputtering, and Ion Beam Sputtering. There are
preferred use cases for each technology, as each has its own strengths and weaknesses.

E-Beam/Thermal Evaporation (Figure 24) is a simple, inexpensive process with a

wide selection of materials and chamber sizes. It works by evaporating the
desired material into a vacuum that then condenses on the optics. It is good for
large parts with lower power density and is generally used for IR or DUV bands.
The stress imparted on the optic can be either tensile or compressive, but it is a
low stress. There coatings have high absorption and low durability, but good
spectral performance.
Figure 24. Depiction of E-Beam coating process.

The Plasma/Ion-Assisted Deposition process (Figure 25) is an extension of the Thermal

Evaporation process, where an Ion Gun is placed in an E-Beam chamber and “assists”
the evaporated particles hit into the optics with higher force. It creates a more dense
coating than Evaporation on its own. It is good for large parts that need a higher laser
power or have stress challenges, but the absorption will be higher than ideal for CW
laser systems.
Figure 25. Depiction of Plasma/Ion-Assisted Evaporation Process.

Page 13 of 18

www.excelitas.com
The DC/RF Magnetron Sputtering process (Figure 26) uses energetic
plasmas to sputter material from targets. It has a high rate of deposition
and large chamber capacities. It is good for general purpose optics, is a
high quality coating and is relatively cheap. It is also a durable coating,
but has higher than desirable absorption for CW laser systems and it can
be challenging to achieve great surface quality with these coatings.
Figure 26. Depiction of DC/RF Magnetron Sputtering Process.

The Ion Beam Sputtering coating process (Figure 27) uses a high-energy ion beam
to sputter material from targets. It produces a very dense, high purity film that is
optimal for High Energy Laser applications. It has unmatched surface quality and
durability characteristics. The downside is that the coating imparts a large amount
of Compressive stress onto the optic, so back side stress compensating coatings are
common. This poses an issue with lightweighted optics, as there is not a coplanar
surface to coat, so generally these types of optic must be pre-figured using MRF
technology to compensate for stress. IBS Coatings have the highest power density
and is optimal for the most critical components. Figure 27. Depiction of Ion Beam Sputtering Process.

IBS is a largely automated method, so once the process is refined and established, it is highly reproducible. In contrast,
evaporative methods require frequent changes in the coating chamber environment and control processes, such as
source material replenishment, making it more difficult achieve a consistent HEL optic. See below in Figure 28 for a
comparison of typical property values from each coating process.

Typical Values
Property
E-Beam/Thermal Evaporation Plasma/Ion-Assisten Dep DC/RF Magnetron Sputtering Ion Beam Sputtering
Accuracy/Control 1% 0.5% 0.2% 0.1%

Deposition Rate 0.7nm/s 0.5nm/s 0.5nm/s 0.1nm/s

Stress 50MPa (T or C) 100MPa (C) 100MPa (C) 400Mpa (C)

Surface Quality 20-10 10-5 10-5 “0-0” (5/ 2 x 0.016)

Stability/Durability Low Medium High Highest

Absorption 100ppm 20ppm 20ppm 2ppm

Figure 28. Typical Process values for various coating processes.

The main takeaway is that while there are many coating technologies available, IBS Coating is the differentiating
technology that enables optics to achieve performance levels required for the highest power High Energy Laser systems.

Page 14 of 18

www.excelitas.com
4.4 Metrology
Assurance of ultra-reliable HEL optics requires the metrology to ensure process optimization as well as the components’
compliance to specification. The Excelitas team is highly proficient in all aspects of optical component and assembly
metrology and are able to certify meeting final requirements internally. For additional information, refer to the cited
publications posted on the Excelitas website, which detail Excelitas and REO’s efforts in developing metrology processes,
specifically defects and laser damage metrology for high-energy laser optics.

REO developed its own internal automated LiDT Testing System that produces detailed information on damage events,
including spatial location and sizes Quality [4, 7, 5, 8]. We are working on developing non-destructive LDT certification
using a Dark Field Microscope and statistical models. We are working to develop a better surface quality certification
method using Cavity Ringdown, TIS and Autmated Surface Quality Inspection (ASQI) tools.

REO IBS Coating capabilities combined with our high sensitivity metrology techniques enable the lowest coating
absorption possible, further enhancing reliability and very high laser damage resistance. We have metrology tools that
can take measurements at 1064, 940, 633, 532 and 355nm wavelengths. In addition to industry standard optical and
spectral instrumentation, below is a partial list of in-house testing we regularly perform:

• In-house laser damage testing systems optimizing coating and surface processes
• Defect-to-damage correlation
• Automated Defect Measurements with unique enhanced diagnostics
• High Sensitivity Absorption Testing (Photothermal Common-Path Interferometry and calorimetry)
• Surface Metrology: Phase and white light interferometry, stylus profilometry, phase contrast microscopy (DIC)
• Optical Loss Measurements: Cavity ringdown at multiple laser wavelengths, custom measurement capabilities
• Automated exposure and data collection
• Spectral Measurements: UV to LWIR spectrophotometry, laser-based
photometry
• Group Delay Dispersion
• Thin Film thickness and index
• Optical System Testing
• Multi-sensor Coordinate Measuring Capabilities

Figure 29. Example of Induced Laser Defect Mechanisms Metrology.

Page 15 of 18

www.excelitas.com
5 Tradeoffs
A major consideration for system design is to understand that an “ideal optic” is effectively impossible to manufacture.
There are necessary tradeoffs that occur as part of the optical manufacturing process that limit the system level
performance. Each specification is a function of many variables, and occasionally there are unintended consequences of
trying to “optimize” for a single performance factor. With so many tradeoffs, it is best to discuss system needs as early as
possible in the design phase with your manufacturing partner to ensure the desired design outcomes are not constraining
your options or increasing your cost. Many times a simpler and more elegant solution can be found if the design intent is
understood from the beginning.

In the below examples, the Blue arrow means a specification “gets better”, the yellow arrow means a specification “gets
worse”, and the grey arrow means there is a “side-effect improvement” due to the combination of factors. These
examples attempt to describe the complex interaction between the various optical specifications.

5.1 Throughput
When prioritizing Throughput for an optic (Figure 30), we will want to have the lowest Absorption and highest laser
damage resistance possible. This will lead to a very stable and durable coating, so the reliability will improve. We also will
want the highest Reflectance and Aperture size possible. This means we will want the best quality films and to have a
large diameter optic. This will generally make Surface Figure harder to control and the system will need to be larger.

The difficult figure control will cause the

beam quality to degrade, but as
absorption is improved, the stability of
the optic will improve because there will
be fewer thermal effects. As aperture
size goes up, the cost will increase for all
aspects of the manufacturing process.
Figure 30. Tradeoffs when prioritizing Throughput
5.2 Beam Quality
If Beam Quality is prioritized (Figure 31), we will want the maximum possible aperture size and best Surface Figure
possible. The Surface Figure will be a major cost driver here, as it is hard to maintain on large surfaces. We also will want
to maximize absorption to minimize thermal effects, but will likely trade some Reflection and Transmission performance
to achieve. The cost is the largest negative impact from this, as controlling Surface Figure is one of the more expensive
processes in optical manufacturing.

Figure 31. Tradeoffs when prioritizing Beam Quality.

Page 16 of 18

www.excelitas.com
5.3 SWaP-C
If Size and Cost, are prioritized (Figure 32) we will
want to shrink the aperture size. This will give up
some performance in beam quality, but if we
increase the Reflection and Transmission
performance, the impacts to Throughput are
minimized. The surface quality can be maintained
at a high level due to the smaller aperture size.
Shrinking the aperture but increasing component
requirements can result in an overall cost benefit Figure 32. Tradeoffs when prioritizing SWAP-C.
as larger aperture optics are such a major cost
driver.

5.4 Reliability
If Reliability is prioritized (Figure 33), it will tend
to only make the system perform better. We will
want to prioritize laser damage, surface quality,
and durability. This can be achieved by using the
highest quality film possible. A result of that will
be improved absorption characteristics and
higher throughput, but the cost will increase. Figure 33. Tradeoffs when prioritizing Reliability.

6 Conclusion
It is important to understand that specifying optical component characteristics is key to optimizing laser system
performance. It is not required that designers understand all of the aspects of these tradeoffs and impacts, but
understanding the critical factors is helpful in order to make the most informed decisions. A few minor specification
changes can result in increased performance, lower cost, and a more reliable system.

6.1 About Excelitas Technologies

Excelitas Technologies® Corp. is a photonics technology leader focused on delivering innovative, high-performance,
market-driven solutions to meet the lighting, optronics, detection and optical technology needs of our OEM customers.

Excelitas has proven design and manufacturing capabilities for market leading High Energy Laser optics, coatings and
assemblies. We have an established history in the design and manufacture these items for many of the US primes and
subassembly suppliers in the High Energy Laser market. Our products are trusted to perform by the most demanding
High Energy Laser users in the United States.

Serving a vast array of applications across biomedical, scientific, safety, security, consumer products, semiconductor,
industrial manufacturing, defense and aerospace sectors, Excelitas stands committed to enabling our customers’ success
in their end-markets. Our photonics team consists of 7,000 professionals working across North America, Europe and Asia,
to serve customers worldwide.

Page 17 of 18

www.excelitas.com
Works Cited
[1] "Lawrence Livermore National Labs," 2020. [Online]. Available: https://lasers.llnl.gov/.
[2] Northrop Grumman, "Northrop Grumman," 19 Nov 2020. [Online]. Available:
https://www.northropgrumman.com/space/solid-state-high-energy-laser-systems/.
[3] Glen P. Perram, S. . J. Cusumano, R. L. Hengehold and S. T. Fiorino, Introduction to Laser Weapon Systems, New
Mexico: The Directed Energy Professional Society, 2010.
[4] T. Turner, Q. Turchette and A. R. Martin, "A statistical study of the relationship between surface quality and laser
induced damage," in SPIE, Boulder, 2012.
[5] A. R. Martin, S. Richman, Q. Turchette and T. Turner, "Large-area defect mapping for laser damage prediction," in
SPIE, Boulder, 2017.
[6] Q. Turchette and T. Turner, "Developing a more useful surface quality metric for laser optics," in SPIE, Boulder,
2011.
[7] S. Richman, A. R. Martin, Q. Turchette and T. Turner, "Adaptive characterization of laser damage from sparse
defects," in SPIE, Boulder, 2014.
[8] S. Richman, A. R. Martin, Q. Turchette and T. Turner, "Method for studying laser-induced damage from sparse
defects," in SPIE, Boulder, 2012.
[9] D. Howland, J. Arenberg, M. Thomas, T. Turner, J. Bellum, C. W. Carr, M. Brophy, A. Krisiloff, N. Carlie and M.
Dabney, "Laser Damage, a New US Standard, What is in it for me?," in SPIE, San Francisco, 2020.
[10] A. Streater and T. Turner, "Trends in Laser Damage," Photonics Spectra, Feb 2007.
[11] R. I. Miller, Laser Weapon Development at Redstone Arsenal, 1960-2015, New Mexico: The Directed Energy
Professional Society, 2016.

Enabling the future through light.

©2021 Excelitas Technologies Corp. All rights reserved. The Excelitas logo and design are registered trademarks of Excelitas Technologies Corp. All other trademarks not owned by Excelitas Technologies or its subsidiaries
that are depicted herein are the property of their respective owners. Excelitas reserves the right to change this document at any time without notice and disclaims liability for editorial, pictorial or typographical errors.

Page 18 of 18

www.excelitas.com