Optical Coatings
Optical Coatings
Optical Coatings
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Fundamental Optics 1<br />
Introduction 1.2<br />
Paraxial Formulas 1.3<br />
Imaging Properties of Lens Systems 1.6<br />
Lens Combination Formulas 1.8<br />
Performance Factors 1.11<br />
Lens Shape 1.17<br />
Lens Combinations 1.18<br />
Diffraction Effects 1.20<br />
Lens Selection 1.23<br />
Spot Size 1.26<br />
Aberration Balancing 1.27<br />
Definition of Terms 1.29<br />
Paraxial Lens Formulas 1.32<br />
Principal-Point Locations 1.36<br />
1.1 1<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong>
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Introduction<br />
Even though several thousand different optical components<br />
are listed in this catalog, performing a few simple calculations will<br />
usually determine the appropriate optics for an application or, at<br />
the very least, narrow the list of choices.<br />
The process of solving virtually any optical engineering problem<br />
can be broken down into two main steps. First, paraxial calculations<br />
(first order) are made to determine critical parameters such<br />
as magnification, focal length(s), clear aperture (diameter), and<br />
object and image position. These paraxial calculations are covered<br />
in the next section of this chapter.<br />
Second, actual components are chosen based on these paraxial<br />
values, and their actual performance is evaluated with special<br />
attention paid to the effects of aberrations. A truly rigorous<br />
performance analysis for all but the simplest optical systems<br />
generally requires computer ray tracing, but simple generalizations<br />
can be used, especially when the lens selection process is<br />
confined to a limited range of component shapes.<br />
In practice, the second step may reveal conflicts with design<br />
constraints, such as component size, cost, or product availability.<br />
System parameters may therefore require modification.<br />
Because some of the terms used in this chapter may not be<br />
familiar to all readers, a glossary of terms is provided beginning<br />
on page 1.29.<br />
Finally, it should be noted that the discussion in this chapter<br />
relates only to systems with uniform illumination; optical systems<br />
for Gaussian beams are covered in Chapter 2, Gaussian Beam<br />
Optics.<br />
ENGINEERING SUPPORT<br />
Melles Griot maintains a staff of knowledgeable,<br />
experienced applications engineers at each of our<br />
facilities worldwide. The information given in this<br />
chapter is sufficient to enable the user to select the<br />
most appropriate catalog lenses for the most<br />
commonly encountered applications. However, when<br />
additional optical engineering support is required,<br />
our applications engineers are available to provide<br />
assistance. Do not hesitate to contact us for help in<br />
product selection or to obtain more detailed<br />
specifications on Melles Griot products.<br />
THE OPTICAL<br />
ENGINEERING PROCESS<br />
Determine basic system<br />
parameters, such as<br />
magnification and<br />
object/image distances<br />
Using paraxial formulas<br />
and known parameters,<br />
solve for remaining values<br />
Pick lens components<br />
based on paraxially<br />
derived values<br />
Determine if chosen<br />
component values conflict<br />
with any basic<br />
system constraints<br />
Estimate performance<br />
characteristics of system<br />
Determine if performance<br />
characteristics meet<br />
original design goals<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
1.2 1 Visit Us OnLine! www.mellesgriot.com
Paraxial Formulas<br />
SIGN CONVENTIONS<br />
The validity of the paraxial lens formulas is dependent on adherence to the following sign conventions:<br />
For lenses: (refer to figure 1.1)<br />
s is 1 for object to left of H<br />
(the first principal point)<br />
s is 5 for object to right of H<br />
s″ is 1 for image to right of H″<br />
(the second principal point)<br />
s″ is 5 for image to left of H″<br />
m is 1 for an inverted image<br />
m is 5 for an upright image<br />
For mirrors:<br />
When using the thin-lens approximation, simply refer to the left and right of the lens.<br />
Figure 1.1<br />
h<br />
front focal point<br />
f object v<br />
H H″<br />
F<br />
s<br />
f<br />
principal points<br />
f<br />
f is 1 for convex (diverging) mirrors<br />
f is 5 for concave (converging) mirrors<br />
s is 1 for object to left of H<br />
s is 5 for object to right of H<br />
s″ is 5 for image to right of H″<br />
s″ is 1 for image to left of H″<br />
m is 1 for an inverted image<br />
m is 5 for an upright image<br />
rear focal point<br />
Note location of object and image relative to front and rear focal points.<br />
f = lens diameter<br />
m =<br />
s″/s = h″/h = magnification or<br />
conjugate ratio, said to be infinite if<br />
either s″ or s is infinite<br />
v = arcsin (f/2s)<br />
h = object height<br />
h″ = image height<br />
Sign conventions<br />
F″<br />
s″<br />
image<br />
s = object distance, positive for object (whether real<br />
or virtual) to the left of principal point H<br />
s″ = image distance (s and s″ are collectively called<br />
conjugate distances, with object and image in<br />
conjugate planes), positive for image (whether real<br />
or virtual) to the right of the principal point H″<br />
f = effective focal length (EFL) which may be positive<br />
(as shown) or negative. f represents both FH and<br />
H″F″, assuming lens to be surrounded by medium<br />
of index 1.0<br />
h″<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.3
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Typically, the first step in optical problem solving is to select a<br />
system focal length based on constraints such as magnification or<br />
conjugate distances (object and image distance). The relationship<br />
among focal length, object position, and image position is<br />
given by<br />
1<br />
f<br />
= 1 s<br />
1<br />
+ .<br />
s′′<br />
m = s ′′<br />
= h ′′ .<br />
(1.2)<br />
s h<br />
This relationship can be used to recast the first formula into the<br />
following forms:<br />
f = m (s + s ′′ )<br />
2<br />
(m + 1)<br />
f = sm<br />
m + 1<br />
s + s′′<br />
f =<br />
m + 2 + 1 m<br />
s (m + 1) = s + s′′<br />
1<br />
= 1 1<br />
4<br />
s′′<br />
f s<br />
1 1 1<br />
= 4<br />
s′′<br />
50 200<br />
s ′′ = 66.7 mm<br />
m = s ′′<br />
= 66.7<br />
s 200 = 0.33<br />
(or real image is 0.33 mm high and inverted).<br />
(1.1)<br />
This formula is referenced to figure 1.1 and the sign conventions<br />
given on page 1.3.<br />
By definition, magnification is the ratio of image size to object<br />
size or<br />
where (s + s″) is the approximate object-to-image distance.<br />
(1.3)<br />
(1.4)<br />
(1.5)<br />
(1.6)<br />
With a real lens of finite thickness, the image distance, object<br />
distance, and focal length are all referenced to the principal points,<br />
not to the physical center of the lens. By neglecting the distance<br />
between the lens’ principal points, known as the hiatus, s + s″<br />
becomes the object-to-image distance. This simplification, called the<br />
thin-lens approximation, can speed up calculation when dealing<br />
with simple optical systems.<br />
Example 1: Object outside Focal Point<br />
A 1-mm-high object is placed on the optical axis, 200 mm left of the<br />
left principal point of a 01 LDX 103 (f = 50 mm). Where is the<br />
image formed, and what is the magnification? (See figure 1.2.)<br />
object<br />
Figure 1.2<br />
Figure 1.3<br />
F1<br />
200 66.7<br />
F 2<br />
image<br />
Example 1 (f = 50 mm, s = 200 mm, s″ = 66.7 mm)<br />
Example 2: Object inside Focal Point<br />
The same object is placed 30 mm left of the left principal point of<br />
the same lens. Where is the image formed, and what is the magnification?<br />
(See figure 1.3.)<br />
1<br />
= 1 1<br />
4<br />
s′′<br />
50 30<br />
s ′′ = 475 mm<br />
m = s ′′ 475<br />
= = 42.5<br />
s 30<br />
(or virtual image is 2.5 mm high and upright).<br />
In this case, the lens is being used as a magnifier, and the image can<br />
be viewed only back through the lens.<br />
image<br />
F 1 F 2<br />
object<br />
Example 2 (f = 50 mm, s = 30 mm, s″ = 475 mm)<br />
Example 3: Object at Focal Point<br />
A 1-mm-high object is placed on the optical axis, 50 mm left of the<br />
first principal point of an 01 LDK 019 (f = 50 mm). Where is the<br />
image formed, and what is the magnification? (See figure 1.4.)<br />
1 1 1<br />
=<br />
s′′<br />
450<br />
4 50<br />
s ′′ = 425 mm<br />
m = s ′′ 425<br />
= = 40.5<br />
s 50<br />
(or virtual image is 0.5 mm high and upright).<br />
1.4 1 Visit Us OnLine! www.mellesgriot.com
object<br />
F 2<br />
Figure 1.4<br />
image<br />
Example 3 (f = 450 mm, s = 50 mm, s″ = 425 mm)<br />
A simple graphical method can also be used to determine paraxial<br />
image location and magnification. This graphical approach relies on<br />
two simple properties of an optical system. First, a ray that enters<br />
the system parallel to the optical axis crosses the optical axis at the<br />
focal point. Second, a ray that enters the first principal point of the<br />
system exits the system from the second principal point parallel to<br />
its original direction (i.e., its exit angle with the optical axis is the same<br />
as its entrance angle). This method has been applied to the three<br />
previous examples illustrated in figures 1.2 through 1.4. Note that by<br />
using the thin-lens approximation, this second property reduces to the<br />
statement that a ray passing through the center of the lens is undeviated.<br />
F-NUMBER AND NUMERICAL APERTURE<br />
The paraxial calculations used to determine necessary element<br />
diameter are based on the concepts of focal ratio (f-number or f/#)<br />
and numerical aperture (NA). The f-number is the ratio of the focal<br />
length of the lens to its clear aperture (effective diameter).<br />
f-number = f f .<br />
(1.7)<br />
To visualize the f-number, consider a lens with a positive focal<br />
length illuminated uniformly with collimated light. The f-number<br />
defines the angle of the cone of light leaving the lens which ultimately<br />
forms the image. This is an important concept when the throughput<br />
or light-gathering power of an optical system is critical, such as<br />
when focusing light into a monochromator or projecting a highpower<br />
image.<br />
The other term used commonly in defining this cone angle is<br />
numerical aperture. Numerical aperture is the sine of the angle made<br />
by the marginal ray with the optical axis. By referring to<br />
figure 1.5 and using simple trigonometry, it can be seen that<br />
f<br />
NA = sin v = (1.8)<br />
2f<br />
or<br />
1<br />
NA =<br />
2(f-number) . (1.9)<br />
F 1<br />
f<br />
2<br />
Figure 1.5<br />
principal surface<br />
F-number and numerical aperture<br />
Ray f-numbers can also be defined for any arbitrary ray if its<br />
conjugate distance and the diameter at which it intersects the<br />
principal surface of the optical system are known.<br />
NOTE<br />
Because the sign convention given previously is not<br />
used universally in all optics texts, the reader may<br />
notice differences in the paraxial formulas. However,<br />
results will be correct as long as a consistent set of<br />
formulas and sign conventions is used.<br />
f<br />
v<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.5
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Imaging Properties of Lens Systems<br />
THE OPTICAL INVARIANT<br />
To understand the importance of the numerical aperture, consider<br />
its relation to magnification. Referring to figure 1.6,<br />
f<br />
NA (object side) = sin v = 2s<br />
f<br />
NA " (image side) = sin v ′′ = 2s ′′<br />
which can be rearranged to show<br />
and<br />
f = 2s sinv<br />
f = 2s ′′ sinv′′<br />
leading to<br />
s′′<br />
s<br />
= sin v<br />
sinv′′<br />
= NA .<br />
NA"<br />
Since s ′′<br />
is simply the magnification of the system,<br />
s<br />
we arrive at<br />
m = NA .<br />
NA"<br />
(1.10)<br />
(1.11)<br />
(1.12)<br />
(1.13)<br />
(1.14)<br />
(1.15)<br />
The magnification of the system is therefore equal to the ratio<br />
of the numerical apertures on the object and image sides of the<br />
system. This powerful and useful result is completely independent<br />
of the specifics of the optical system, and it can often be used to determine<br />
the optimum lens diameter in situations involving aperture<br />
constraints.<br />
When a lens or optical system is used to create an image of a<br />
source, it is natural to assume that, by increasing the diameter (f)<br />
of the lens, we will be able to collect more light and thereby produce<br />
a brighter image. However, because of the relationship between<br />
magnification and numerical aperture, there can be a theoretical limit<br />
beyond which increasing the diameter has no effect on lightcollection<br />
efficiency or image brightness.<br />
Since the numerical aperture of a ray is given by f/2s, once a<br />
focal length and magnification have been selected, the value of NA<br />
sets the value of f. Thus, if one is dealing with a system in which the<br />
numerical aperture is constrained on either the object or image<br />
side, increasing the lens diameter beyond this value will increase<br />
system size and cost but will not improve performance (i.e., throughput<br />
or image brightness). This concept is sometimes referred to as<br />
the optical invariant.<br />
Example: System with Fixed Input NA<br />
Two very common applications of simple optics involve coupling<br />
light into an optical fiber or into the entrance slit of a monochromator.<br />
Although these problems appear to be quite different, they<br />
both have the same limitation — they have a fixed numerical<br />
aperture. For monochromators, this limit is usually expressed in<br />
terms of the f-number. In addition to the fixed numerical aperture,<br />
they both have a fixed entrance pupil (image) size.<br />
Suppose it is necessary, using a singlet lens from this catalog, to<br />
couple the output of an incandescent bulb with a filament 1 mm in<br />
diameter into an optical fiber as shown in figure 1.7. Assume that the<br />
fiber has a core diameter of 100 mm and a numerical aperture of 0.25,<br />
and that the design requires that the total distance from the source<br />
to the fiber be 110 mm. Which lenses are appropriate?<br />
By definition, the magnification must be 0.1. Letting s + s″ total<br />
110 mm (using the thin-lens approximation), we can use equation<br />
1.3,<br />
f = m (s + s ′′ )<br />
(m + 1) 2<br />
to determine that the focal length is 9.1 mm. To determine the<br />
conjugate distances, s and s″, we utilize equation 1.6,<br />
s (m + 1) = s + s ′′,<br />
and find that s = 100 mm and s″ = 10 mm.<br />
We can now use the relationship NA = Ω/2s or NA″ = Ω/2s″ to<br />
derive Ω, the optimum clear aperture (effective diameter) of the lens.<br />
With an image numerical aperture of 0.25 and an image distance<br />
(s″) of 10 mm,<br />
f<br />
0.25 = 20<br />
f<br />
= 5 mm.<br />
Accomplishing this imaging task with a single lens therefore<br />
requires an optic with a 9.1-mm focal length and a 5-mm diameter.<br />
Using a larger diameter lens will not result in any greater system<br />
throughput because of the limited input numerical aperture of the<br />
optical fiber. The singlet lenses in this catalog that meet these criteria<br />
are 01 LPX 003, which is plano-convex, and 01 LDX 003 and<br />
01 LDX 005, which are biconvex.<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
SAMPLE CALCULATION<br />
To understand how to use this relationship between magnification<br />
and numerical aperture, consider the following example.<br />
Making some simple calculations has reduced our choice of<br />
lenses to just three. Chapter 2, Gaussian Beam Optics, discusses<br />
how to make a final choice of lenses based on various performance<br />
criteria.<br />
1.6 1 Visit Us OnLine! www.mellesgriot.com
Figure 1.6<br />
Figure 1.7<br />
f<br />
f<br />
2<br />
object side<br />
magnification =<br />
h"<br />
=<br />
0.1<br />
= 0.1X<br />
h 1.0<br />
filament<br />
h = 1 mm<br />
v<br />
s<br />
Numerical aperture and magnification<br />
s + s" = 110 mm<br />
f<br />
NA = = 0.025<br />
2s<br />
s = 100 mm<br />
optical system<br />
f = 9.1 mm<br />
f = 5 mm<br />
f<br />
NA" = = 0.25<br />
2s"<br />
fiber core<br />
h" = 0.1 mm<br />
s" = 10 mm<br />
<strong>Optical</strong> system geometry for focusing the output of an incandescent bulb into an optical fiber<br />
s″<br />
v″<br />
image side<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.7
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Lens Combination Formulas<br />
PARAXIAL LENS COMBINATION FORMULAS<br />
Many optical tasks require several lenses in order to achieve an<br />
acceptable level of performance. One possible approach to lens<br />
combinations is to consider each image formed by each lens as the<br />
object for the next lens and so on. This is a valid approach, but it is<br />
time consuming and unnecessary.<br />
It is much simpler to calculate the effective (combined) focal<br />
length and principal-point locations and then use these results in<br />
any subsequent paraxial calculations (see figure 1.8). They can even<br />
be used in the optical invariant calculations described in the<br />
preceding section.<br />
EFFECTIVE FOCAL LENGTH<br />
The following formulas show how to calculate the effective focal<br />
length and principal-point locations for a combination of any two<br />
arbitrary components. The approach for more than two lenses is very<br />
simple: calculate the values for the first two elements, then perform<br />
the same calculation for this combination with the next lens. This is<br />
continued until all lenses in the system are accounted for.<br />
The expression for the combination focal length is the same<br />
whether lens separation distances are large or small and whether f 1<br />
and f 2 are positive or negative:<br />
f =<br />
1<br />
f<br />
ff 1 2<br />
f 1 + f2 4 .<br />
d<br />
This may be more familiar in the form<br />
s ′′ =<br />
2<br />
= 1 f<br />
+<br />
1<br />
f<br />
4 d<br />
ff<br />
.<br />
1 2 1 2<br />
f 2 (f1<br />
4 d)<br />
.<br />
f + f 4 d<br />
1 2<br />
z = s 2′′ 4 f .<br />
(1.16)<br />
(1.17)<br />
Notice that the formula is symmetric with respect to interchange<br />
of the lenses (end-for-end rotation of the combination) at constant<br />
d. The next two formulas are not.<br />
COMBINATION FOCAL-POINT LOCATION<br />
For all cases,<br />
COMBINATION SECONDARY<br />
PRINCIPAL-POINT LOCATION<br />
(1.18)<br />
Because the thin-lens approximation is obviously highly invalid<br />
for most combinations, the ability to determine the location of the<br />
secondary principal point is vital for accurate determination of d when<br />
another element is added. The simplest formula for this calculates<br />
how far the secondary principal point of the final (second) element<br />
is moved by being part of the combination:<br />
(1.19)<br />
COMBINATION EXAMPLES<br />
It is possible for a lens combination or system to exhibit principal<br />
planes that are far removed from the system. When such systems<br />
are themselves combined, negative values of d may occur. Probably<br />
the simplest example of a negative d-value situation is shown in<br />
figure 1.9. Meniscus lenses with steep surfaces have external principal<br />
planes. When two of these lenses are brought into contact, a<br />
negative value of d can occur. Other combined-lens examples are<br />
shown in figures 1.10 through 1.13.<br />
SYMBOLS<br />
f<br />
f 1<br />
f 2<br />
d<br />
= combination focal length (EFL), positive if<br />
combination final focal point falls to right of<br />
combination secondary principal point,<br />
negative otherwise.<br />
= focal length (EFL) of first element.<br />
= focal length (EFL) of second element.<br />
= distance from secondary principal point of<br />
first element to primary principal point of<br />
second element (positive if primary principal<br />
point is to right of the secondary principal<br />
point, negative otherwise).<br />
s 2 ″ = distance from secondary principal point of<br />
second element to final combination focal<br />
point (location of final image for object at<br />
infinity to left), positive if the focal point is<br />
to right of second element secondary principal<br />
point.<br />
z<br />
= distance to combination secondary principal<br />
point measured from secondary principal<br />
point of second element, positive if<br />
combination secondary principal point is to<br />
right of secondary principal point of second<br />
element.<br />
Note: These paraxial formulas apply to coaxial<br />
combinations of both thick and thin lenses immersed<br />
in any fluid with refractive index independent of<br />
position. They assume that light propagates from left<br />
to right through an optical system.<br />
1.8 1 Visit Us OnLine! www.mellesgriot.com
Figure 1.8<br />
Figure 1.9<br />
INDIVIDUAL ELEMENT<br />
1st element<br />
COMBINATION<br />
2 elements<br />
SUBSYSTEM<br />
d<br />
d<br />
2nd element<br />
z, from formula<br />
3rd element<br />
combination secondary principal plane<br />
(to find combination primary principal plane,<br />
apply procedure to reversed combination<br />
resulting from end-to-end rotation)<br />
subsystem secondary principal plane<br />
n-1 elements nth element to be added to complete the system<br />
d<br />
z, from formula<br />
system secondary<br />
COMPLETE SYSTEM<br />
principal plane<br />
principal planes<br />
not “crossed”<br />
system primary principal plane (secondary principal<br />
plane located by z formula for reversed system)<br />
lens combinations or systems may exhibit “crossed” principal planes; single lenses cannot<br />
SUBSYSTEM<br />
principal planes internal but “crossed”<br />
n-1 elements<br />
Generalization from combinations to systems<br />
1 2 3 4<br />
d<br />
subsystem secondary principal plane<br />
nth element to be added to complete the system<br />
subsystem primary principal plane<br />
3 4 1 2<br />
d>0 d
Fundamental Optics<br />
d<br />
s 2 ″<br />
f 1 f 2<br />
z<br />
f
Performance Factors<br />
After paraxial formulas have been used to select values for component<br />
focal length(s) and diameter(s), the final step is to select<br />
actual lenses. As in any engineering problem, this selection process<br />
involves a number of tradeoffs, including performance, cost, weight,<br />
and environmental factors.<br />
The performance of real optical systems is limited by several<br />
factors, including lens aberrations and light diffraction. The magnitude<br />
of these effects can be calculated with relative ease.<br />
Numerous other factors, such as lens manufacturing tolerances<br />
and component alignment, impact the performance of an optical<br />
system. Although these are not considered explicitly in the following<br />
discussion, it should be kept in mind that if calculations indicate that<br />
a lens system only just meets the desired performance criteria, in<br />
practice it may fall short of this performance as a result of other<br />
factors. In critical applications, it is generally better to select a lens<br />
whose calculated performance is significantly better than needed.<br />
DIFFRACTION<br />
Diffraction, a natural property of light arising from its wave<br />
nature, poses a fundamental limitation on any optical system. Diffraction<br />
is always present, although its effects may be masked if<br />
the system has significant aberrations. When an optical system is<br />
essentially free from aberrations, its performance is limited solely<br />
by diffraction, and it is referred to as diffraction limited.<br />
In calculating diffraction, we simply need to know the focal<br />
length(s) and aperture diameter(s); we do not consider other lensrelated<br />
factors such as shape or index of refraction.<br />
Since diffraction increases with increasing f-number, and aberrations<br />
decrease with increasing f-number, determining optimum<br />
system performance often involves finding a point where the combination<br />
of these factors has a minimum effect.<br />
ABERRATIONS<br />
To determine the precise performance of a lens system, we can<br />
trace the path of light rays through it, using Snell’s law at each<br />
optical interface to determine the subsequent ray direction. This<br />
process, called ray tracing, is usually accomplished on a computer.<br />
When this process is completed, it is typically found that not all<br />
the rays pass through the points or positions predicted by paraxial<br />
theory. These deviations from ideal imaging are called lens<br />
aberrations.<br />
The direction of a light ray after refraction at the interface between<br />
two homogeneous, isotropic media of differing index of refraction is<br />
given by Snell’s law:<br />
n 1 sinß 1 = n 2 sinß 2<br />
( 1.20)<br />
where ß 1 is the angle of incidence, ß 2 is the angle of refraction, and<br />
both angles are measured from the surface normal as shown in figure<br />
1.14.<br />
material 1<br />
index n 1<br />
material 2<br />
index n2<br />
Figure 1.14<br />
wavelength l d<br />
Technical Assistance<br />
v 1<br />
v 2<br />
Refraction of light at a dielectric boundary<br />
APPLICATION NOTE<br />
Detailed performance analysis of an optical system<br />
is accomplished using computerized ray-tracing<br />
software. Melles Griot applications engineers have<br />
the capability to provide a ray-tracing analysis of<br />
simple catalog components systems. If you need<br />
assistance in determining the performance of your<br />
optical system, or in selecting optimum components<br />
for your particular application, please contact your<br />
nearest Melles Griot office.<br />
Alternately, a database containing prescription<br />
information for most of the components listed in this<br />
catalog is available on the catalog CD-ROM. If you<br />
would like to obtain a copy of this database, please<br />
contact your Melles Griot representative.<br />
For analysis of more complex optical systems,<br />
or the design of totally custom lenses, Melles Griot<br />
<strong>Optical</strong> Systems, located in Rochester, New York, can<br />
supply the necessary support. This group specializes<br />
in the design and fabrication of high-precision,<br />
multielement lens systems. For more information<br />
about their capabilities, please call your Melles Griot<br />
representative.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.11
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Even though tools for precise analysis of an optical system are<br />
becoming easier to use and are readily available, it is still quite useful<br />
to have a method for quickly estimating lens performance. This<br />
not only saves time in the initial stages of system specification, but<br />
can also help achieve a better starting point for any further<br />
computer optimization.<br />
The first step in developing these rough guidelines is to realize<br />
that the sine functions in Snell’s law can be expanded in an infinite<br />
Taylor series:<br />
3<br />
1 1 1<br />
5<br />
1<br />
sin v = v 4 v /3! + v /5! 4 v /7! + v /9! 4. . .<br />
The first approximation we can make is to replace all sine functions<br />
with their arguments (i.e., replace sin ß 1 with ß 1 itself and so<br />
on). This is called first-order or paraxial theory because only the first<br />
terms of the sine expansions are used. Design of any optical system<br />
generally starts with this approximation using the paraxial formulas.<br />
The assumption that sinß = ß is reasonably valid for ß close to zero<br />
(i.e., high f-number lenses). With more highly curved surfaces (and<br />
particularly marginal rays), paraxial theory yields increasingly large<br />
deviations from real performance because sinß ≠ ß. These deviations<br />
are known as aberrations. Because a perfect optical system (one<br />
without any aberrations) would form its image at the point and to<br />
the size indicated by paraxial theory, aberrations are really a measure<br />
of how the image differs from the paraxial prediction.<br />
As already stated, exact ray tracing is the only rigorous way to<br />
analyze real lens surfaces. Before the advent of computers, this was<br />
excessively tedious and time consuming. Seidel addressed this issue<br />
by developing a method of calculating aberrations resulting from<br />
the ß 1 3 /3! term. The resultant third-order lens aberrations are therefore<br />
called Seidel aberrations.<br />
To simplify these calculations, Seidel put the aberrations of an<br />
optical system into several different classifications. In monochromatic<br />
light they are spherical aberration, astigmatism, field<br />
curvature, coma, and distortion. In polychromatic light there are<br />
also chromatic aberration and lateral color. Seidel developed<br />
methods to approximate each of these aberrations without actually<br />
tracing large numbers of rays using all the terms in the sine<br />
expansions.<br />
In actual practice, aberrations occur in combinations rather<br />
than alone. This system of classifying them, which makes analysis<br />
much simpler, gives a good description of optical system image<br />
quality. In fact, even in the era of powerful ray-tracing software,<br />
Seidel’s formula for spherical aberration is still widely used.<br />
7<br />
1<br />
9<br />
1<br />
SPHERICAL ABERRATION<br />
Figure 1.15 illustrates how an aberration-free lens focuses<br />
incoming collimated light. All rays pass through the focal point F ″.<br />
The lower figure shows the situation more typically encountered in<br />
single lenses. The farther from the optical axis the ray enters the<br />
lens, the nearer to the lens it focuses (crosses the optical axis). The<br />
distance along the optical axis between the intercept of the rays<br />
that are nearly on the optical axis (paraxial rays) and the rays that<br />
go through the edge of the lens (marginal rays) is called longitudinal<br />
spherical aberration (LSA). The height at which these rays<br />
intercept the paraxial focal plane is called transverse spherical<br />
aberration (TSA). These quantities are related by<br />
TSA = LSA ! tan u″.<br />
aberration-free lens<br />
u″<br />
paraxial focal plane<br />
longitudinal spherical aberration<br />
LSA<br />
F″<br />
F″<br />
TSA<br />
transverse spherical aberration<br />
Figure 1.15 Spherical aberration of a plano-convex lens<br />
(1.21)<br />
Spherical aberration is dependent on lens shape, orientation, and<br />
conjugate ratio, as well as on the index of refraction of the materials<br />
present. Parameters for choosing the best lens shape and orientation<br />
for a given task are presented later in this chapter. However, the<br />
1.12 1 Visit Us OnLine! www.mellesgriot.com
third-order, monochromatic, spherical aberration of a plano-convex<br />
lens used at infinite conjugate ratio can be estimated by<br />
spot size due to spherical aberration = 0.067 f . (1.22)<br />
f/# 3<br />
Theoretically, the simplest way to eliminate or reduce spherical<br />
aberration is to make the lens surface(s) with a varying radius of curvature<br />
(i.e., an aspheric surface) designed to exactly compensate for<br />
the fact that sin v ≠ v at larger angles. In practice, however, most lenses<br />
with high surface quality are manufactured by grinding and polishing<br />
techniques that naturally produce spherical or cylindrical surfaces.<br />
The manufacture of aspheric surfaces is more complex, and it is<br />
difficult to produce a lens of sufficient surface accuracy to eliminate<br />
spherical aberration completely. Fortunately, these aberrations<br />
can be virtually eliminated, for a chosen set of conditions, by combining<br />
the effects of two or more spherical (or cylindrical) surfaces.<br />
In general, simple positive lenses have undercorrected spherical<br />
aberration, and negative lenses usually have overcorrected spherical<br />
aberration. By combining a positive lens made from low-index glass<br />
with a negative lens made from high-index glass, it is possible to produce<br />
a combination in which the spherical aberrations cancel but<br />
the focusing powers do not. The simplest examples of this are<br />
cemented doublets, such as the 01 LAO series which produce<br />
minimal spherical aberration when properly used.<br />
object point<br />
Figure 1.16<br />
optical axis<br />
tangential plane<br />
tangential image<br />
(focal line)<br />
principal ray<br />
sagittal plane<br />
optical system<br />
Astigmatism represented by sectional views<br />
ASTIGMATISM<br />
When an off-axis object is focused by a spherical lens, the natural<br />
asymmetry leads to astigmatism. The system appears to have two<br />
different focal lengths.<br />
As shown in figure 1.16, the plane containing both optical axis<br />
and object point is called the tangential plane. Rays that lie in this<br />
plane are called tangential rays. Rays not in this plane are referred<br />
to as skew rays. The chief, or principal, ray goes from the object<br />
point through the center of the aperture of the lens system. The<br />
plane perpendicular to the tangential plane that contains the principal<br />
ray is called the sagittal or radial plane.<br />
The figure illustrates that tangential rays from the object come<br />
to a focus closer to the lens than do rays in the sagittal plane. When<br />
the image is evaluated at the tangential conjugate, we see a line in<br />
the sagittal direction. A line in the tangential direction is formed at<br />
the sagittal conjugate. Between these conjugates, the image is either<br />
an elliptical or a circular blur. Astigmatism is defined as the<br />
separation of these conjugates.<br />
The amount of astigmatism in a lens depends on lens shape only<br />
when there is an aperture in the system that is not in contact with the<br />
lens itself. (In all optical systems there is an aperture or stop, although<br />
in many cases it is simply the clear aperture of the lens element itself.)<br />
Astigmatism strongly depends on the conjugate ratio.<br />
paraxial<br />
focal plane<br />
sagittal image (focal line)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.13
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
COMA<br />
In spherical lenses, different parts of the lens surface exhibit different<br />
degrees of magnification. This gives rise to an aberration<br />
known as coma. As shown in figure 1.17, each concentric zone of<br />
a lens forms a ring-shaped image called a comatic circle. This causes<br />
blurring in the image plane (surface) of off-axis object points. An<br />
off-axis object point is not a sharp image point, but it appears as a<br />
characteristic comet-like flare. Even if spherical aberration is<br />
corrected and the lens brings all rays to a sharp focus on axis, a<br />
lens may still exhibit coma off axis. See figure 1.18.<br />
As with spherical aberration, correction can be achieved by<br />
using multiple surfaces. Alternatively, a sharper image may be<br />
produced by judiciously placing an aperture, or stop, in an optical<br />
system to eliminate the more marginal rays.<br />
FIELD CURVATURE<br />
Even in the absence of astigmatism, there is a tendency of optical<br />
systems to image better on curved surfaces than on flat planes. This<br />
effect is called field curvature (see figure 1.19). In the presence of astigmatism,<br />
this problem is compounded because there are two separate<br />
astigmatic focal surfaces that correspond to the tangential and<br />
sagittal conjugates.<br />
Field curvature varies with the square of field angle or the square<br />
of image height. Therefore, by reducing the field angle by one-half,<br />
it is possible to reduce the blur from field curvature to a value of 0.25<br />
of its original size.<br />
S<br />
1′<br />
1<br />
1<br />
1′<br />
1<br />
1′<br />
P,O<br />
Figure 1.18<br />
Figure 1.19<br />
points on lens<br />
1<br />
4 2<br />
1′<br />
4′ 2′<br />
3 3′ 0 3′ 3<br />
2′ 4′<br />
1′<br />
2<br />
4<br />
1<br />
positive transverse coma<br />
focal plane<br />
Positive transverse coma<br />
spherical focal surface<br />
Field curvature<br />
corresponding<br />
points on S<br />
1<br />
4<br />
1′<br />
2<br />
4′ 2′<br />
3<br />
3′<br />
S<br />
60°<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 1.17<br />
Imaging an off-axis point source by a lens with positive transverse coma<br />
1.14 1 Visit Us OnLine! www.mellesgriot.com
Positive lens elements usually have inward curving fields, and negative<br />
lenses have outward curving fields. Field curvature can thus<br />
be corrected to some extent by combining positive and negative<br />
lens elements.<br />
DISTORTION<br />
The image field not only may have curvature but may also be<br />
distorted. The image of an off-axis point may be formed at a<br />
location on this surface other than that predicted by the simple<br />
paraxial equations. This distortion is different from coma (where<br />
rays from an off-axis point fail to meet perfectly in the image<br />
plane). Distortion means that even if a perfect off-axis point image<br />
is formed, its location on the image plane is not correct. Furthermore,<br />
the amount of distortion usually increases with increasing<br />
image height. The effect of this can be seen as two different kinds<br />
of distortion: pincushion and barrel (see figure 1.20). Distortion<br />
does not lower system resolution; it simply means that the image<br />
shape does not correspond exactly to the shape of the object.<br />
Distortion is a separation of the actual image point from the<br />
paraxially predicted location on the image plane and can be<br />
expressed either as an absolute value or as a percentage of the<br />
paraxial image height.<br />
It should be apparent that a lens or lens system has opposite<br />
types of distortion depending on whether it is used forward or backward.<br />
This means that if a lens were used to make a photograph,<br />
and then used in reverse to project it, there would be no distortion<br />
in the final screen image. Also, perfectly symmetrical optical systems<br />
at 1:1 magnification have no distortion or coma.<br />
CHROMATIC ABERRATION<br />
The aberrations previously described are purely a function of the<br />
shape of the lens surfaces, and can be observed with monochromatic<br />
light. There are, however, other aberrations that arise when<br />
these optics are used to transform light containing multiple<br />
wavelengths.<br />
OBJECT<br />
PINCUSHION<br />
DISTORTION<br />
BARREL<br />
DISTORTION<br />
The index of refraction of a material is a function of wavelength.<br />
Known as dispersion, this is discussed in Chapter 4, Material<br />
Properties. From Snell’s law (see equation 1.20), it can be seen that<br />
light rays of different wavelengths or colors will be refracted at<br />
different angles since the index is not a constant. Figure 1.21 shows<br />
the result when polychromatic collimated light is incident on a positive<br />
lens element. Because the index of refraction is higher for<br />
shorter wavelengths, these are focused closer to the lens than the<br />
longer wavelengths. Longitudinal chromatic aberration is defined<br />
as the axial distance from the nearest to the farthest focal point.<br />
As in the case of spherical aberration, positive and negative<br />
elements have opposite signs of chromatic aberration. Once again,<br />
by combining elements of nearly opposite aberration to form a<br />
doublet, chromatic aberration can be partially corrected. It is necessary<br />
to use two glasses with different dispersion characteristics,<br />
so that the weaker negative element can balance the aberration of<br />
the stronger, positive element.<br />
Variations of Aberrations with Aperture,<br />
Field Angle, and Image Height<br />
Aberration<br />
Lateral Spherical<br />
Longitudinal Spherical<br />
Coma<br />
Astigmatism<br />
Field Curvature<br />
Distortion<br />
Chromatic<br />
white light ray<br />
Aperture<br />
blue light ray<br />
blue focal point<br />
red light ray<br />
red focal point<br />
Figure 1.20 Pincushion and barrel distortion Figure 1.21 Longitudinal chromatic aberration<br />
(Ω)<br />
Ω 3<br />
Ω 2<br />
Ω 2<br />
Ω<br />
Ω<br />
—<br />
—<br />
Field Angle<br />
(ß)<br />
—<br />
—<br />
ß<br />
ß 2<br />
ß 2<br />
ß 3<br />
—<br />
Image Height<br />
(y)<br />
—<br />
—<br />
y<br />
y 2<br />
y 2<br />
y 3<br />
—<br />
longitudinal<br />
chromatic<br />
aberration<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.15
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
LATERAL COLOR<br />
Lateral color is the difference in image height between blue and<br />
red rays. Figure 1.22 shows the chief ray of an optical system<br />
consisting of a simple positive lens and a separate aperture. Because<br />
of the change in index with wavelength, blue light is refracted more<br />
strongly than red light, which is why rays intercept the image plane<br />
at different heights. Stated simply, magnification depends on color.<br />
Lateral color is very dependent on system stop location.<br />
For many optical systems, the third-order term is all that may<br />
be needed to quantify aberrations. However, in highly corrected<br />
systems or in those having large apertures or a large angular field<br />
of view, third-order theory is inadequate. In these cases, exact ray<br />
tracing is absolutely essential.<br />
aperture<br />
Figure 1.22<br />
Lateral color<br />
red light ray<br />
blue light ray<br />
focal plane<br />
lateral color<br />
APPLICATION NOTE<br />
Achromatic Doublets Are Superior<br />
to Simple Lenses<br />
Because achromatic doublets correct for spherical<br />
as well as chromatic aberration, they are often<br />
superior to simple lenses for focusing collimated<br />
light or collimating point sources, even in purely<br />
monochromatic light.<br />
Although there is no simple formula that can be<br />
used to estimate the spot size of a doublet, the<br />
tables on page 1.26 give sample values that can be<br />
used to estimate the performance of other catalog<br />
achromats.<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
1.16 1 Visit Us OnLine! www.mellesgriot.com
Lens Shape<br />
Aberrations described in the preceding section are highly<br />
dependent on application, lens shape, and material of the lens (or,<br />
more exactly, its index of refraction). The singlet shape that minimizes<br />
spherical aberration at a given conjugate ratio is called best-form.<br />
The criterion for best-form at any conjugate ratio is that the marginal<br />
rays are equally refracted at each of the lens/air interfaces. This<br />
minimizes the effect of sin v ≠ v. It is also the criterion for minimum<br />
surface-reflectance loss. Another benefit is that absolute coma is<br />
nearly minimized for best-form shape, at both infinite and unit<br />
conjugate ratios.<br />
To further explore the dependence of aberrations on lens shape, it<br />
is helpful to make use of the Coddington shape factor, q, defined as<br />
q = (r 2 + r 1 ) . (1.23)<br />
(r 4 r )<br />
Figure 1.23<br />
2 1<br />
Figure 1.23 shows the transverse and longitudinal spherical<br />
aberration of a singlet lens as a function of the shape factor, q. In this<br />
particular instance, the lens has a focal length of 100 mm, operates<br />
at f/5, has an index of refraction of 1.518722 (BK7 at the mercury<br />
green line, 546.1 nm), and is being operated at the infinite conjugate<br />
ratio. It is also assumed that the lens itself is the aperture stop. An<br />
asymmetric shape that corresponds to a q-value of about 0.7426 for<br />
this material and wavelength is the best singlet shape for on-axis<br />
imaging. Best-form shapes are used in Melles Griot laser-line-focusing<br />
singlet lenses. It is important to note that the best-form shape is<br />
dependent on refractive index. For example, with a high-index<br />
material, such as silicon, the best-form lens for the infinite conjugate<br />
ratio is a meniscus shape.<br />
ABERRATIONS IN MILLIMETERS<br />
5<br />
4<br />
3<br />
2<br />
1<br />
42<br />
exact transverse spherical<br />
aberration (TSA)<br />
41.5 41 40.5 0 0.5 1 1.5 2<br />
SHAPE FACTOR (q)<br />
exact longitudinal spherical aberration (LSA)<br />
Aberrations of positive singlets at infinite conjugate ratio as a function of shape<br />
At infinite conjugate with a typical glass singlet, the plano-convex<br />
shape (q = 1), with convex side toward the infinite conjugate, performs<br />
nearly as well as the best-form lens. Because a plano-convex lens costs<br />
much less to manufacture than an asymmetric biconvex singlet, these<br />
lenses are quite popular. Furthermore, this lens shape exhibits nearminimum<br />
total transverse aberration and near-zero coma when used<br />
off axis, thus enhancing its utility.<br />
For imaging at unit magnification (s = s″ = 2f), a similar analysis<br />
would show that a symmetric biconvex lens is the best shape. Not<br />
only is spherical aberration minimized, but coma, distortion, and<br />
lateral chromatic aberration exactly cancel each other out. These<br />
results are true regardless of material index or wavelength, which<br />
explains the utility of symmetric convex lenses, as well as symmetrical<br />
optical systems in general. However, if a remote stop is present,<br />
these aberrations may not cancel each other quite as well.<br />
For wide-field applications, the best-form shape is definitely not<br />
the optimum singlet shape, especially at the infinite conjugate ratio,<br />
since it yields maximum field curvature. The ideal shape is determined<br />
by the situation and may require rigorous ray-tracing analysis.<br />
It is possible to achieve much better correction in an optical system<br />
by using more than one element. The cases of an infinite<br />
conjugate ratio system and a unit conjugate ratio system are<br />
discussed in the following section.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.17
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Lens Combinations<br />
INFINITE CONJUGATE RATIO<br />
As shown in the previous discussion, the best-form singlet lens<br />
for use at infinite conjugate ratios is generally nearly plano-convex.<br />
Figure 1.24 shows a plano-convex lens (01 LPX 023) with<br />
incoming collimated light at a wavelength of 546.1 nm. This drawing,<br />
including the rays traced through it, is shown to exact scale. The<br />
marginal ray (ray f-number 1.5) strikes the paraxial focal plane significantly<br />
off the optical axis.<br />
This situation can be improved by using a two-element system.<br />
The second part of the figure shows a precision achromat (01 LAO 014),<br />
which consists of a positive low-index (crown glass) element cemented<br />
to a negative meniscus high-index (flint glass) element. This is drawn<br />
to the same scale as the plano-convex lens. No spherical aberration<br />
can be discerned in the lens. Of course, not all of the rays pass exactly<br />
through the paraxial focal point; however, in this case, the departure<br />
is measured in micrometers, rather than in millimeters, as in the case<br />
of the plano-convex lens. Additionally, chromatic aberration (not<br />
shown) is much better corrected in the doublet. Even though these<br />
lenses are known as achromatic doublets, it is important to remember<br />
that even with monochromatic light the doublet’s performance is<br />
superior.<br />
Figure 1.24 also shows the f-number at which singlet performance<br />
becomes unacceptable. The ray with f-number 7.5 practically intercepts<br />
the paraxial focal point, and the f/3.8 ray is fairly close. This useful<br />
drawing, which can be scaled to fit a plano-convex lens of any focal<br />
length, can be used to estimate the magnitude of its spherical aberration,<br />
although lens thickness affects results slightly.<br />
UNIT CONJUGATE RATIO<br />
Figure 1.25 shows three possible systems for use at the unit<br />
conjugate ratio. All are shown to the same scale and using the<br />
same ray f-numbers with a light wavelength of 546.1 nm. The first<br />
system is a symmetric biconvex lens (01 LDX 027), the best-form<br />
singlet in this application. Clearly, significant spherical aberration<br />
is present in this lens at f/2.7. Not until f/13.3 does the ray closely<br />
approach the paraxial focus.<br />
A dramatic improvement in performance is gained by using two<br />
identical plano-convex lenses with convex surfaces facing and nearly<br />
in contact. Those shown in figure 1.25 are both 01 LPX 081. The combination<br />
of these two lenses yields almost exactly the same focal<br />
length as the biconvex lens. To understand why this configuration<br />
improves performance so dramatically, consider that if the biconvex<br />
lens were split down the middle, we would have two identical<br />
plano-convex lenses, each working at an infinite conjugate ratio,<br />
but with the convex surface toward the focus. This orientation is<br />
opposite to that shown to be optimum for this shape lens. On the other<br />
hand, if these lenses are reversed, we have the system just described<br />
but with a better correction of the spherical aberration.<br />
ray f-numbers<br />
1.5<br />
1.9<br />
2.5<br />
3.8<br />
7.5<br />
1.5<br />
1.9<br />
2.5<br />
3.8<br />
7.5<br />
PLANO-CONVEX LENS<br />
ACHROMAT<br />
paraxial image plane<br />
01 LPX 023<br />
01 LAO 014<br />
Figure 1.24 Single-element plano-convex lens compared<br />
with a two-element achromat<br />
The previous examples indicate that an achromat is superior in<br />
performance to a singlet when used at the infinite conjugate ratio<br />
and at low f-numbers. Since the unit conjugate case can be thought<br />
of as two lenses, each working at the infinite conjugate ratio, the next<br />
step is to replace the plano-convex singlets with achromats, yielding<br />
a four-element system. The third part of figure 1.25 shows a system<br />
composed of two 01 LAO 037 lenses. Once again, spherical aberration<br />
is not evident, even in the f/2.7 ray.<br />
1.18 1 Visit Us OnLine! www.mellesgriot.com
ay f-numbers<br />
2.7<br />
3.3<br />
4.4<br />
6.7<br />
13.3<br />
SYMMETRIC BICONVEX LENS<br />
IDENTICAL PLANO-CONVEX LENSES<br />
2.7<br />
3.3<br />
4.4<br />
6.7<br />
13.3<br />
2.7<br />
3.3<br />
4.4<br />
6.7<br />
13.3<br />
01 LDX 027<br />
01 LPX 081<br />
IDENTICAL ACHROMATS<br />
01 LAO 037<br />
Figure 1.25 Three possible systems for use at the unit conjugate ratio<br />
paraxial image plane<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.19
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Diffraction Effects<br />
In all light beams, some energy is spread outside the region predicted<br />
by rectilinear propagation. This effect, known as diffraction,<br />
is a fundamental and inescapable physical phenomenon.<br />
Diffraction can be understood by considering the wave nature<br />
of light. Huygen’s principle (figure 1.26) states that each point on<br />
a propagating wavefront is an emitter of secondary wavelets. The<br />
combined focus of these expanding wavelets forms the propagating<br />
wave. Interference between the secondary wavelets gives rise to a<br />
fringe pattern that rapidly decreases in intensity with increasing<br />
angle from the initial direction of propagation. Huygen’s principle<br />
nicely describes diffraction, but rigorous explanation demands a<br />
detailed study of wave theory.<br />
Diffraction effects are traditionally classified into either Fresnel<br />
or Fraunhofer types. Fresnel diffraction is primarily concerned<br />
with what happens to light in the immediate neighborhood of a<br />
diffracting object or aperture. It is thus only of concern when the<br />
illumination source is close to this aperture or object. Consequently,<br />
Fresnel diffraction is rarely important in most optical setups.<br />
Fraunhofer diffraction, however, is often very important. This is<br />
the light-spreading effect of an aperture when the aperture (or<br />
object) is illuminated with an infinite source (plane-wave illumination)<br />
and the light is sensed at an infinite distance (far-field) from<br />
this aperture.<br />
From these overly simple definitions, one might assume that<br />
Fraunhofer diffraction is important only in optical systems with<br />
infinite conjugate, whereas Fresnel diffraction equations should be<br />
considered at finite conjugate ratios. Not so. A lens or lens system<br />
of finite positive focal length with plane-wave input maps the farfield<br />
diffraction pattern of its aperture onto the focal plane; therefore,<br />
it is Fraunhofer diffraction that determines the limiting<br />
performance of optical systems. More generally, at any conjugate<br />
ratio, far-field angles are transformed into spatial displacements<br />
in the image plane.<br />
APPLICATION NOTE<br />
Rayleigh Criterion<br />
In imaging applications, spatial resolution is ultimately<br />
limited by diffraction. Calculating the maximum possible<br />
spatial resolution of an optical system requires an<br />
arbitrary definition of what is meant by resolving two<br />
features. In the Rayleigh criterion, it is assumed that<br />
two separate point sources can be resolved when the<br />
center of the Airy disc from one overlaps the first<br />
dark ring in the diffraction pattern of the second. In<br />
this case, the smallest resolvable distance, d, is<br />
0.61 l<br />
d = = 1.22 l f/# .<br />
N.A.<br />
CIRCULAR APERTURE<br />
Fraunhofer diffraction at a circular aperture dictates the<br />
fundamental limits of performance for circular lenses. It is important<br />
to remember that the spot size, caused by diffraction, of a circular<br />
lens is<br />
d = 2.44 l f/#<br />
(1.24)<br />
where d is the diameter of the focused spot produced from planewave<br />
illumination and l is the wavelength of light being focused.<br />
Notice that it is the f-number of the lens, not its absolute diameter,<br />
that determines this limiting spot size.<br />
The diffraction pattern resulting from a uniformly illuminated circular<br />
aperture actually consists of a central bright region, known as<br />
the Airy disc (see figure 1.27), which is surrounded by a number of much<br />
fainter rings. Each ring is separated by a circle of zero intensity. The<br />
irradiance distribution in this pattern can be described by<br />
I = I<br />
x 0<br />
Figure 1.26<br />
secondary<br />
wavelets<br />
wavefront<br />
aperture<br />
⎡ 2J 1(x)<br />
⎤<br />
⎢<br />
⎣ x<br />
⎥<br />
⎦<br />
where I 0 = peak irradiance in image<br />
J (x) = x ( 41)<br />
1<br />
Huygen’s principle<br />
2<br />
J (x) = Bessel function of the first kind of order unity<br />
1<br />
x =<br />
∞<br />
∑<br />
n=1<br />
πD sin v<br />
l<br />
n+1<br />
2n4<br />
2<br />
x<br />
(n 4 1)!n!2<br />
2n4<br />
1<br />
where l = wavelength<br />
D= aperture diameter<br />
v = angular radius from pattern maximum.<br />
some light diffracted<br />
into this region<br />
wavefront<br />
(1.25)<br />
This useful formula shows the far-field irradiance distribution from<br />
a uniformly illuminated circular aperture of diameter, D.<br />
1.20 1 Visit Us OnLine! www.mellesgriot.com
AIRY DISC DIAMETER = 2.44 l f/#<br />
Figure 1.27 Center of a typical diffraction pattern for a<br />
circular aperture<br />
SLIT APERTURE<br />
A slit aperture, which is mathematically simpler, is useful in<br />
relation to cylindrical optical elements. The irradiance distribution<br />
in the diffraction pattern of a uniformly illuminated slit aperture is<br />
described by<br />
where<br />
I = I<br />
x 0<br />
I<br />
0<br />
x =<br />
Energy Distribution in the Diffraction Pattern of a Circular or Slit Aperture<br />
Ring or Band<br />
⎡sin x ⎤<br />
⎢ ⎥<br />
⎣⎢<br />
x ⎦⎥<br />
Central Maximum<br />
First Dark<br />
First Bright<br />
Second Dark<br />
Second Bright<br />
Third Dark<br />
Third Bright<br />
Fourth Dark<br />
Fourth Bright<br />
Fifth Dark<br />
Note: Position variable (x) is defined in the text.<br />
2<br />
= peak irradiance in image<br />
pw sin v<br />
l<br />
where l = wavelength<br />
w = slit width<br />
v = angular deviation from pattern maximum.<br />
Position<br />
(x)<br />
0.0<br />
1.22p<br />
1.64p<br />
2.23p<br />
2.68p<br />
3.24p<br />
3.70p<br />
4.24p<br />
4.71p<br />
5.24p<br />
(1.26)<br />
Circular Aperture<br />
Relative<br />
Intensity<br />
(I x /I 0 )<br />
1.0<br />
0.0<br />
0.0175<br />
0.0<br />
0.0042<br />
0.0<br />
0.0016<br />
0.0<br />
0.0008<br />
0.0<br />
ENERGY DISTRIBUTION TABLE<br />
The table below shows the major features of pure (unaberrated)<br />
Fraunhofer diffraction patterns of circular and slit apertures. The<br />
table shows the position, relative intensity, and percentage of total<br />
pattern energy corresponding to each ring or band. It is especially<br />
convenient to characterize positions in either pattern with the same<br />
variable x. This variable is related to field angle in the circular<br />
aperture case by<br />
l<br />
sin v = x pD<br />
l<br />
sin v = x pw<br />
where w is the slit width, p has its usual meaning, and D, w, and l<br />
are all in the same units (preferably millimeters).<br />
Linear instead of angular field positions are simply found from<br />
r = s″ tan (v)<br />
(1.29)<br />
where s″ is the secondary conjugate distance. This last result is often<br />
seen in a different form, namely the diffraction-limited spot-size<br />
equation. For a circular lens that was stated at the outset of this<br />
section:<br />
d = 2.44 l f/#<br />
Energy<br />
in Ring<br />
(%)<br />
83.8<br />
7.2<br />
2.8<br />
1.5<br />
1.0<br />
Position<br />
(x)<br />
0.0<br />
1.00p<br />
1.43p<br />
2.00p<br />
2.46p<br />
3.00p<br />
3.47p<br />
4.00p<br />
4.48p<br />
5.00p<br />
Slit Aperture<br />
Relative<br />
Intensity<br />
(I x /I 0 )<br />
1.0<br />
0.0<br />
0.0472<br />
0.0<br />
0.0165<br />
0.0<br />
0.0083<br />
0.0<br />
0.0050<br />
0.0<br />
(1.27)<br />
where D is the aperture diameter. For a slit aperture, this relationship<br />
is given by<br />
(1.28)<br />
(see 1.24)<br />
This value represents the smallest spot size that can be achieved<br />
by an optical system with a circular aperture of a given f-number.<br />
Energy<br />
in Band<br />
(%)<br />
90.3<br />
4.7<br />
1.7<br />
0.8<br />
0.5<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.21
Fundamental Optics<br />
The graph in figure 1.28 shows the form of both circular and slit<br />
aperture diffraction patterns when plotted on the same normalized<br />
scale. Aperture diameter is equal to slit width so that patterns between<br />
x-values and angular deviations in the far-field are the same.<br />
when dealing with Gaussian beams, the location of the focused spot<br />
also departs from that predicted by the paraxial equations given<br />
in this chapter. This is also detailed in chapter 2.<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
GAUSSIAN BEAMS<br />
Apodization, or nonuniformity of aperture irradiance, alters<br />
diffraction patterns. If pupil irradiance is nonuniform, the formulas<br />
and results given previously do not apply. This is important to<br />
remember because most laser-based optical systems do not have<br />
uniform pupil irradiance. The output beam of a laser operating<br />
in the TEM 00 mode has a smooth Gaussian irradiance profile.<br />
Formulas to determine the focused spot size from such a beam are<br />
discussed in Chapter 2, Gaussian Beam Optics. Furthermore,<br />
NORMALIZED PATTERN IRRADIANCE (y)<br />
CIRCULAR APERTURE<br />
1.0<br />
.9<br />
.8<br />
.7<br />
.6<br />
.5<br />
.4<br />
.3<br />
.2<br />
.1<br />
slit<br />
aperture<br />
91.0% within first bright ring<br />
83.9% in Airy disc<br />
0.0<br />
48 47 46 45 44 43 42 41 0 1 2 3 4 5 6 7 8<br />
POSITION IN IMAGE PLANE (x)<br />
90.3% in<br />
central maximum<br />
95.0% within the two<br />
adjoining subsidiary maxima<br />
circular<br />
aperture<br />
y =<br />
c<br />
⎛ 2J 1(x)<br />
⎞<br />
⎜<br />
⎝ x<br />
⎟<br />
⎠<br />
2<br />
∞<br />
x<br />
n+1<br />
where J 1(x) = x ∑( 4 1)<br />
(n 4 1)!n!2<br />
n=1<br />
Note : J (x) is the Bessel function<br />
1<br />
2n 4 2<br />
of the first kind of order unity.<br />
2n 4 1<br />
p<br />
x = D sinv<br />
l<br />
l = wavelength<br />
D = aperture diameter<br />
v = angular radius from pattern maximum<br />
2<br />
⎛ sin x ⎞<br />
p<br />
ys<br />
= ⎜ , where x w sin<br />
x<br />
⎟<br />
= v<br />
⎝ ⎠<br />
l<br />
l = wavelength<br />
w = slit width<br />
v = angular deviation direction of pattern<br />
maximum<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
SLIT APERTURE<br />
Figure 1.28 Fraunhofer diffraction pattern of a singlet slit superimposed on the Fraunhofer diffraction pattern of a<br />
circular aperture<br />
1.22 1 Visit Us OnLine! www.mellesgriot.com
Lens Selection<br />
Having discussed the most important factors that affect a lens or<br />
a lens system’s performance, we will now address the practical matter<br />
of selecting the optimum catalog components for a particular task.<br />
The following useful relationships are important to keep in mind<br />
throughout the selection process:<br />
$ Diffraction-limited spot size = 2.44 ¬ f/#<br />
$ Approximate on-axis spot size<br />
of a plano-convex lens at the infinite<br />
conjugate resulting from spherical aberration =<br />
$ <strong>Optical</strong> invariant =<br />
Example 1: Collimating an Incandescent Source<br />
Produce a collimated beam from a quartz halogen bulb having<br />
a 1-mm-square filament. Collect the maximum amount of light<br />
possible and produce a beam with the lowest possible divergence<br />
angle.<br />
This problem, illustrated in figure 1.29, involves the typical tradeoff<br />
between light-collection efficiency and resolution (where a beam<br />
is being collimated rather than focused, resolution is defined by beam<br />
divergence). To collect more light, it is necessary to work at a low<br />
f-number, but because of aberrations, higher resolution (lower divergence<br />
angle) will be achieved by working at a higher f-number.<br />
In terms of resolution, the first thing to realize is that the<br />
minimum divergence angle (in radians) that can be achieved using<br />
any lens system is the source size divided by system focal length. An<br />
off-axis ray (from the edge of the source) entering the first principal<br />
point of the system exits the second principal point at the same<br />
angle. Therefore, increasing system focal length improves this limiting<br />
divergence because the source appears smaller.<br />
An optic that can produce a spot size of 1 mm when focusing a<br />
perfectly collimated beam is therefore required. Since source size is<br />
inherently limited, it is pointless to strive for better resolution. This<br />
level of resolution can be achieved easily with a plano-convex lens.<br />
Figure 1.29<br />
m = NA<br />
NA" .<br />
Collimating an incandescent source<br />
f<br />
0.067 f<br />
f/# 3<br />
While angular divergence decreases with increasing focal length,<br />
spherical aberration of a plano-convex lens increases with increasing<br />
focal length. To determine the appropriate focal length, set the<br />
spherical aberration formula for a plano-convex lens equal to the<br />
source (spot) size:<br />
0.067 f<br />
3<br />
f/#<br />
= 1 mm.<br />
This ensures a lens that meets the minimum performance needed.<br />
To select a focal length, make an arbitrary f-number choice. As<br />
can be seen from the relationship, as we lower the f-number (increase<br />
collection efficiency), we decrease the focal length, which will worsen<br />
the resultant divergence angle (minimum divergence = 1 mm/f).<br />
In this example, we will accept f/2 collection efficiency, which gives<br />
us a focal length of about 120 mm. For f/2 operation we would<br />
need a minimum diameter of 60 mm. The 01 LPX 209 fits this<br />
specification exactly. Beam divergence would be about 8 mrad.<br />
Finally, we need to verify that we are not operating below the<br />
theoretical diffraction limit. In this example, the numbers (1-mm<br />
spot size) indicate that we are not, since<br />
diffraction-limited spot size = 2.44 ! 0.5 mm ! 2 = 2.44 mm.<br />
Example 2: Coupling an Incandescent Source into a Fiber<br />
On pages 1.6 and 1.7 we considered a system in which the output<br />
of an incandescent bulb with a filament of 1 mm in diameter was<br />
to be coupled into an optical fiber with a core diameter of 100 µm<br />
and a numerical aperture of 0.25. From the optical invariant and<br />
other constraints given in the problem, we determined that system<br />
focal length is 9.1 mm, diameter = 5 mm, s = 100 mm, s″ = 10 mm,<br />
NA″ = 0.25, and NA = 0.025 (or f/2 and f/20). The singlet lenses<br />
that match these specifications are the plano-convex 01 LPX 003<br />
or biconvex lenses 01 LDX 003 and 01 LDX 005. The closest<br />
achromat would be the 01 LAO 001.<br />
v min =<br />
source size<br />
f<br />
v min<br />
(see eq. 1.22)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.23
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
We can immediately reject the biconvex lenses because of<br />
spherical aberration. We can estimate the performance of the<br />
01 LPX 003 on the focusing side by using our spherical aberration<br />
formula:<br />
0.067 (10)<br />
spot size = = 84 mm.<br />
3<br />
2<br />
We will ignore, for the moment, that we are not working at the<br />
infinite conjugate.<br />
This is slightly smaller than the 100-µm spot size we’re trying<br />
to achieve. However, since we are not working at infinite conjugate,<br />
the spot size will be larger than given by our simple calculation.<br />
This lens is therefore likely to be marginal in this situation,<br />
especially if we consider chromatic aberration. A better choice is the<br />
achromat. Although a computer ray trace would be required to<br />
determine its exact performance, it is virtually certain to provide adequate<br />
performance.<br />
Example 3: Symmetric Fiber-to-Fiber Coupling<br />
Couple an optical fiber with an 8-µm core and a 0.15 numerical<br />
aperture into another fiber with the same characteristics. Assume<br />
a wavelength of 0.5 µm.<br />
This problem, illustrated in figure 1.30, is essentially a 1:1 imaging<br />
situation. We want to collect and focus at a numerical aperture of<br />
0.15 or f/3.3, and we need a lens with an 8-µm spot size at this<br />
f-number. Based on the lens combination discussion on page 1.8,<br />
our most likely setup is either a pair of identical plano-convex lenses<br />
or achromats, faced front to front. To determine the necessary focal<br />
length for a plano-convex lens, we again use the spherical aberration<br />
estimate formula:<br />
0.067 f<br />
3<br />
3.3<br />
= 0.008 mm.<br />
This formula yields a focal length of 4.3 mm and a minimum<br />
diameter of 1.3 mm. The 01 LPX 423 meets these criteria. The<br />
biggest problem with utilizing these tiny, short focal length lenses<br />
is the practical considerations of handling, mounting, and positioning<br />
them. Since using a pair of longer focal length singlets would<br />
result in unacceptable performance, the next step might be to<br />
use a pair of the slightly longer focal length, larger achromats,<br />
such as the 01 LAO 001. The performance data, given on page 1.26,<br />
shows that this combination does provide the required 8-mm spot<br />
diameter.<br />
Because fairly small spot sizes are being considered here, it is<br />
important to make sure that the system is not being asked to work<br />
below the diffraction limit:<br />
2.44 ! 0.5 mm ! 3.3 = 4 mm .<br />
Since this is half the spot size caused by aberrations, it can be<br />
safely assumed that diffraction will not play a significant role here.<br />
An entirely different approach to a fiber-coupling task such as<br />
this would be a pair of spherical ball lenses (06 LMS series), listed<br />
on page 15.15, or one of the gradient-index lenses (06 LGT series),<br />
listed on page 15.19.<br />
s = f<br />
s"= f<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 1.30<br />
Symmetric fiber-to-fiber coupling<br />
1.24 1 Visit Us OnLine! www.mellesgriot.com
Example 4: Diffraction-Limited Performance<br />
Determine at what f-number a plano-convex lens being used at<br />
an infinite conjugate ratio with 0.5-mm wavelength light becomes<br />
diffraction limited (i.e., the effects of diffraction exceed those caused<br />
by aberration).<br />
To solve this problem, set the equations for diffraction-limited spot<br />
size and third-order spherical aberration equal to each other. The<br />
result depends upon focal length, since aberrations scale with focal<br />
length, while diffraction is solely dependent upon f-number. Substituting<br />
some common focal lengths into this formula, we get f/8.6<br />
at f = 100 mm, f/7.2 at f = 50 mm, and f/4.8 at f = 10 mm.<br />
2.44 0.5 m f/# = 0.067 !<br />
! m !<br />
f<br />
3<br />
f/#<br />
or<br />
1/4<br />
f/# = (54.9 ! f) .<br />
When working with these focal lengths (and under the conditions<br />
previously stated), we can assume essentially diffraction-limited<br />
performance above these f-numbers. Keep in mind, however, that<br />
this treatment does not take into account manufacturing tolerances<br />
or chromatic aberration, which will be present in polychromatic<br />
applications.<br />
MELLES GRIOT LENS DATABASE<br />
A database containing prescription information<br />
for most of the optical components listed in this<br />
catalog is included in the Melles Griot catalog on<br />
CD-ROM. This database, in a Zemax format,<br />
facilitates the determination of<br />
• Spot size<br />
• Prescription information<br />
• Wavefront distortion.<br />
Please contact our sales department for your free<br />
Melles Griot Catalog on CD-ROM:<br />
Phone: 1-800-835-2626 / (949) 261-5600<br />
FAX: (949) 261-7790<br />
E-mail: mglit@irvine.mellesgriot.com<br />
Non-US customers should contact the nearest<br />
Melles Griot office (see back cover).<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.25
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Spot Size<br />
In general, the performance of a lens or lens system in a specific<br />
circumstance should be determined by an exact trigonometric ray<br />
trace. Melles Griot applications engineers can supply ray-trace<br />
data for particular lenses and systems of catalog components on<br />
request. However, for certain situations, some simple guidelines<br />
can be used for lens selection. The optimum working conditions<br />
for some of the lenses in this catalog have already been presented.<br />
The following tables give some quantitative results for a variety<br />
of simple and compound lens systems that can be constructed<br />
from standard catalog optics.<br />
In interpreting these tables, remember that these theoretical values<br />
obtained from computer ray tracing consider only the effects<br />
of ideal geometric optics. Effects of manufacturing tolerances have<br />
not been considered. Furthermore, remember that using more than<br />
one element provides a higher degree of correction but makes<br />
alignment more difficult. When actually choosing a lens or a lens<br />
system, it is important to note the tolerances and specifications<br />
clearly described for each Melles Griot lens in the product listings.<br />
The tables give spot size for a variety of lenses used at several different<br />
f-numbers. All the tables are for on-axis, uniformly illuminated,<br />
collimated input light at 632.8 nm. They assume that the lens is<br />
facing in the direction that produces a minimum spot size. When<br />
the spot size caused by aberrations is smaller or equal to the<br />
diffraction-limited spot size, the notation “DL’’ appears next to<br />
the entry. The shorter focal length lenses produce smaller spot sizes<br />
because aberrations increase linearly as a lens is scaled up.<br />
Focal Length = 10 mm<br />
f/2<br />
f/3<br />
f/5<br />
f/10<br />
01 LDX 005 01 LPX 005 01 LAO 001<br />
550<br />
120<br />
30<br />
15 (DL)<br />
*Diffraction-limited performance is indicated by DL.<br />
Focal Length = 60 mm<br />
Spot Size (µm)*<br />
95<br />
25<br />
8 (DL)<br />
15 (DL)<br />
4<br />
5 (DL)<br />
8 (DL)<br />
15 (DL)<br />
The effect on spot size caused by spherical aberration is strongly<br />
dependent on f-number. For a plano-convex singlet, spherical<br />
aberration is inversely dependent on the cube of the f-number. For<br />
doublets, this relationship can be even higher. On the other hand,<br />
the spot size caused by diffraction increases linearly with f-number.<br />
Thus, for some lens types, spot size at first decreases and then<br />
increases with f-number, meaning that there is some optimum<br />
performance point where both aberrations and diffraction combine<br />
to form a minimum.<br />
Unfortunately, these results cannot be generalized to situations<br />
where the lenses are used off axis. This is particularly true of the<br />
achromat/aplanatic meniscus lens combinations because their<br />
performance degrades rapidly off axis.<br />
Focal Length = 30 mm<br />
f/2<br />
f/3<br />
f/5<br />
f/10<br />
01 LPX 049 01 LAO 024<br />
350<br />
90<br />
17<br />
15 (DL)<br />
*Diffraction-limited performance is indicated by DL.<br />
Spot Size (µm)*<br />
Spot Size (µm)*<br />
80<br />
11<br />
8 (DL)<br />
15 (DL)<br />
01 LAO 059 &<br />
01 LAM 059<br />
4<br />
5 (DL)<br />
8 (DL)<br />
15 (DL)<br />
01 LDX 123 01 LPX 127 01 LAO 079 01 LAO 126 & 01 LAM 126<br />
f/2<br />
f/3<br />
f/5<br />
f/10<br />
800<br />
225<br />
42<br />
15 (DL)<br />
600<br />
200<br />
30<br />
15 (DL)<br />
80<br />
35<br />
9<br />
15 (DL)<br />
6<br />
5 (DL)<br />
8 (DL)<br />
15 (DL)<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
*Diffraction-limited performance is indicated by DL.<br />
1.26 1 Visit Us OnLine! www.mellesgriot.com
Aberration Balancing<br />
To improve system performance, optical designers make sure<br />
that the total aberration contribution from all surfaces taken together<br />
sums to nearly zero. Normally, such a process requires computerized<br />
analysis and optimization. However, there are some simple<br />
guidelines that can be used to achieve this with lenses available in<br />
this catalog. This approach can yield systems that operate at a much<br />
lower f-number than can usually be achieved with simple lenses.<br />
Specifically, we will examine how to null the spherical aberration<br />
from two or more lenses in collimated, monochromatic light. Thus,<br />
this technique will be most useful for laser beam focusing and<br />
expanding.<br />
Figure 1.31 shows the third-order longitudinal spherical<br />
aberration coefficients for four of the most common positive and<br />
negative lens shapes when used with parallel, monochromatic<br />
incident light. The plano-convex and plano-concave lenses both<br />
show minimum spherical aberration when oriented with their curved<br />
surface facing the incident parallel beam. All other configurations<br />
exhibit larger amounts of spherical aberration. With these lens types,<br />
it is now possible to show how various systems can be corrected for<br />
spherical aberration.<br />
A two-element laser beam expander is a good starting example.<br />
In this case, two lenses are separated by a distance which is the<br />
sum of their focal lengths, so that the overall system focal length is<br />
infinite. This system will not focus incoming collimated light, but<br />
it will change the beam diameter. By definition, each of the lenses<br />
is operating at the same f-number.<br />
The equation for longitudinal spherical aberration shows that<br />
for two lenses with the same f-number, aberration varies directly with<br />
the focal lengths of the lenses. The sign of the aberration is the same<br />
as focal length. Thus, it should be possible to correct the spherical<br />
Figure 1.31<br />
positive lenses<br />
negative lenses<br />
aberration<br />
coefficient<br />
(k)<br />
plano-convex (reversed) 01 LPX<br />
plano-concave (reversed) 01 LPK<br />
symmetric-convex 01 LDX<br />
symmetric-concave 01 LDK<br />
longitudinal spherical aberration (3rd order) =<br />
kf<br />
f/# 2<br />
plano-convex (normal) 01 LPX<br />
plano-concave (normal) 01 LPK<br />
1.069 0.403 0.272<br />
Third-order longitudinal spherical aberration of typical lens shapes<br />
aberration of this Galilean-type beam expander, which consists of<br />
a positive focal length objective and a negative diverging lens.<br />
If a plano-convex lens of focal length f 1 oriented in the normal<br />
direction is combined with a plano-concave lens of focal length f 2<br />
oriented in its reverse direction, the total spherical aberration of<br />
the system is<br />
LSA = 0.272 f<br />
2<br />
f/#<br />
f<br />
f<br />
1<br />
2<br />
1<br />
1.069 f2<br />
+ .<br />
2<br />
f/#<br />
After setting this equal to zero, we obtain<br />
1.069<br />
= 4<br />
0.272 = 4 3.93.<br />
(1.30)<br />
To make the magnitude of aberration contributions of the two<br />
elements equal so they will cancel out, and thus correct the system,<br />
select the focal length of the positive element to be 3.93 times that<br />
of the negative element.<br />
Figure 1.32 shows a beam-expander system made up of catalog<br />
elements, in which the focal length ratio is 4:1. This simple system is<br />
corrected to about 1/6 wavelength at 632.8 nm, even though the objective<br />
is operating at f/4 with a 20-mm aperture diameter. This is remarkably<br />
good wavefront correction for such a simple system; one would<br />
normally assume that a doublet objective would be needed and a<br />
complex diverging lens as well. This analysis does not take into<br />
account manufacturing tolerances.<br />
A beam expander of lower magnification can also be derived<br />
from this information. If a symmetric-convex objective is used<br />
together with a reversed plano-concave diverging lens, the aberration<br />
coefficients are in the ratio of 1.069/40.403 = 42.65. Figure 1.32<br />
shows a system of catalog lenses that provides a magnification of<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.27
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
a) CORRECTED 4!BEAM EXPANDER<br />
f= 420 mm<br />
10-mm diameter<br />
plano-concave<br />
01 LPK 001<br />
b) CORRECTED 2.7x BEAM EXPANDER<br />
f= 420 mm<br />
10-mm diameter<br />
plano-concave<br />
01 LPK 001<br />
f= 80 mm<br />
22.4-mm diameter<br />
plano-convex<br />
01 LPX 149<br />
c) SPHERICALLY CORRECTED 25-mm EFL f/2.0 OBJECTIVE<br />
f= 425 mm<br />
25-mm diameter<br />
plano-concave<br />
01 LPK 003<br />
f= 54 mm<br />
32-mm diameter<br />
symmetric-convex<br />
01 LDX 119<br />
f= 50 mm (2)<br />
27-mm diameter<br />
plano-convex<br />
01 LPX 108<br />
2.7 (the closest possible given the available focal lengths). The<br />
maximum wavefront error in this case is only 1/4 wave, even though<br />
the objective is working at f/3.3.<br />
The relatively fast speed of these objectives is a great advantage<br />
in minimizing the length of these beam expanders. They would be<br />
particularly useful with Nd:YAG and argon-ion lasers, which tend<br />
to have large output beam diameters.<br />
These same principles can be utilized to create high numerical<br />
aperture objectives that might be used as laser focusing lenses.<br />
Figure 1.32 shows an objective consisting of an initial negative<br />
element, followed by two identical plano-convex positive elements.<br />
Again, all of the elements operate at the same f-number, so that<br />
their aberration contributions are proportional to their focal lengths.<br />
To obtain zero total spherical aberration from this configuration,<br />
we must satisfy<br />
or<br />
1.069 f + 0.272 f + 0.272 f = 0<br />
f<br />
f<br />
1<br />
2<br />
1 2 2<br />
= 40.51.<br />
Therefore, a corrected system should result if the focal length of<br />
the negative element is just about half that of each of the positive<br />
lenses. In this case, f 1 = 425 mm and f 2 = 50 mm yield a total system<br />
focal length of about 25 mm and an f-number of approximately<br />
f/2. This objective, corrected to 1/6 wave, has the additional advantage<br />
of a very long working distance.<br />
UV OPTICS<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 1.32<br />
balancing<br />
Combining catalog lenses for aberration<br />
The material presented in this section is based on the work of John<br />
F. Forkner.<br />
Melles Griot now offers a selection of UV optics<br />
ranging from 193 to 355 nm. See Chapter 16,<br />
UV Optics, for details.<br />
1.28 1 Visit Us OnLine! www.mellesgriot.com
Definition of Terms<br />
FOCAL LENGTH (f)<br />
Two distinct terms describe the focal lengths associated with<br />
every lens or lens system. The effective focal length (EFL) or<br />
equivalent focal length (denoted f in figure 1.33) determines<br />
magnification and hence the image size. The term f appears<br />
frequently in the lens formulas and tables of standard lenses.<br />
Unfortunately, f is measured with reference to principal points<br />
which are usually inside the lens so the meaning of f is not<br />
immediately apparent when a lens is visually inspected.<br />
The second type of focal length relates the focal plane positions<br />
directly to landmarks on the lens surfaces (namely the vertices)<br />
which are immediately recognizable. It is not simply related to image<br />
size but is especially convenient for use when one is concerned about<br />
correct lens positioning or mechanical clearances. Examples of this<br />
second type of focal length are the front focal length (FFL, denoted<br />
f f in figure 1.33) and the back focal length (BFL, denoted f b ).<br />
The convention in all of the figures (with the exception of a single<br />
deliberately reversed ray) is that light travels from left to right.<br />
secondary principal surface<br />
primary principal point<br />
t<br />
secondary principal point<br />
e<br />
primary principal surface<br />
ray from object at infinity<br />
rear (secondary)<br />
focal point<br />
ray from object at infinity<br />
r<br />
primary vertex A<br />
2<br />
optical axis<br />
1<br />
F<br />
H H″<br />
A 2 secondary vertex<br />
F″<br />
front (primary)<br />
r 1<br />
focal point<br />
reversed ray locates front focal<br />
point or primary principal surface<br />
front focal point<br />
A = front focus to front<br />
edge distance<br />
B = rear edge to rear<br />
focus distance<br />
Figure 1.33 Focal length and focal points<br />
f f<br />
A<br />
f<br />
f = effective focal length;<br />
may be positive (as shown)<br />
or negative<br />
f f = front focal length<br />
f b = back focal length<br />
t c<br />
FOCAL POINT (F OR F″)<br />
Rays that pass through or originate at either focal point must be,<br />
on the opposite side of the lens, parallel to the optical axis. This<br />
fact is the basis for locating both focal points.<br />
PRIMARY PRINCIPAL SURFACE<br />
Let us imagine that rays originating at the front focal point F (and<br />
therefore parallel to the optical axis after emergence from the opposite<br />
side of the lens) are singly refracted at some imaginary surface,<br />
instead of twice refracted (once at each lens surface) as actually<br />
happens. There is a unique imaginary surface, called the principal<br />
surface, at which this can happen.<br />
To locate this unique surface, consider a single ray traced from<br />
the air on one side of the lens, through the lens and into the air on<br />
the other side. The ray is broken into three segments by the lens.<br />
Two of these are external (in the air), and the third is internal (in<br />
the glass). The external segments can be extended to a common<br />
point of intersection (certainly near, and usually within, the lens). The<br />
t e = edge thickness<br />
t c = center thickness<br />
f b<br />
B<br />
f<br />
rear focal point<br />
r 1 = radius of curvature of first<br />
surface (positive if center of<br />
curvature is to right)<br />
r 2 = radius of curvature of second<br />
surface (negative if center of<br />
curvature is to left)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.29
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
principal surface is the locus of all such points of intersection of<br />
extended external ray segments. The principal surface of a perfectly<br />
corrected optical system is a sphere centered on the focal point.<br />
Near the optical axis, the principal surface is nearly flat, and<br />
for this reason, it is sometimes referred to as the principal plane.<br />
SECONDARY PRINCIPAL SURFACE<br />
This term is defined analogously to the primary principal surface,<br />
but it is used for a collimated beam incident from the left and focused<br />
to the rear focal point F ≤ on the right. Rays in that part of the<br />
beam nearest the axis can be thought of as once refracted at the<br />
secondary principal surface, instead of being refracted by both lens<br />
surfaces.<br />
PRIMARY PRINCIPAL POINT (H)<br />
OR FIRST NODAL POINT<br />
This point is the intersection of the primary principal surface with<br />
the optical axis.<br />
SECONDARY PRINCIPAL POINT (H≤)<br />
OR SECONDARY NODAL POINT<br />
This point is the intersection of the secondary principal surface<br />
with the optical axis.<br />
CONJUGATE DISTANCES (S AND S″)<br />
The conjugate distances are the object distance, s, and image<br />
distance, s″. Specifically, s is the distance from the object to H, and<br />
s″ is the distance from H″ to the image location. The term infinite<br />
conjugate ratio refers to the situation in which a lens is either focusing<br />
incoming collimated light, or being used to collimate a source (therefore<br />
either s or s″ is infinity).<br />
PRIMARY VERTEX (A 1 )<br />
The primary vertex is the intersection of the primary lens surface<br />
with the optical axis.<br />
SECONDARY VERTEX (A 2 )<br />
The secondary vertex is the intersection of the secondary lens<br />
surface with the optical axis.<br />
BACK FOCAL LENGTH (f b )<br />
This length is the distance from the secondary vertex (A 2 ) to<br />
the rear focal point (F″ ).<br />
EDGE-TO-FOCUS DISTANCES (A AND B)<br />
A is the distance from the front focal point to the front edge of<br />
the lens. B is the distance from the rear edge of the lens to the rear<br />
focal point. Both distances are presumed always to be positive.<br />
REAL IMAGE<br />
A real image is one in which the light rays actually converge;<br />
if a screen were placed at the point of focus, an image would be<br />
formed on it.<br />
VIRTUAL IMAGE<br />
A virtual image does not represent an actual convergence of light<br />
rays. A virtual image can be viewed only by looking back through<br />
the optical system, such as in the case of a magnifying glass.<br />
F-NUMBER (F/#)<br />
The f-number (also known as the focal ratio, relative aperture,<br />
or speed) of a lens system is defined to be the effective focal length<br />
divided by system clear aperture. Ray f-number is the conjugate<br />
distance for that ray divided by the height at which it intercepts the<br />
principal surface.<br />
f/# = f φ . (1.31)<br />
NUMERICAL APERTURE (NA)<br />
The numerical aperture of a lens system is defined to be the sine<br />
of the angle, v 1 , that the marginal ray (the ray that exits the lens<br />
system at its outer edge) makes with the optical axis multiplied by<br />
the index of refraction (n) of the medium. The numerical aperture<br />
can be defined for any ray as the sine of the angle made by that ray<br />
with the optical axis multiplied by the index of refraction:<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
EFFECTIVE FOCAL LENGTH (EFL, f)<br />
Assuming that the lens is surrounded by air or vacuum (refractive<br />
index 1.0), this is both the distance from the front focal point (F) to the<br />
primary principal point (H) and the distance from the secondary principal<br />
point (H″) to the rear focal point (F″). Later we use f to designate<br />
the paraxial effective focal length for the design wavelength (¬ 0 ).<br />
FRONT FOCAL LENGTH (f f )<br />
This length is the distance from the front focal point (F) to the<br />
primary vertex (A 1 ).<br />
NA = n sin v.<br />
MAGNIFICATION POWER<br />
Often, positive lenses intended for use as simple magnifiers are<br />
rated with a single magnification, such as 4#. To create a virtual<br />
image for viewing with the human eye, in principle, any positive<br />
lens can be used at an infinite number of possible magnifications.<br />
However, there is usually a narrow range of magnifications that<br />
will be comfortable for the viewer. Typically, when the viewer adjusts<br />
the object distance so that the image appears to be essentially at<br />
1.30 1 Visit Us OnLine! www.mellesgriot.com
infinity (which is a comfortable viewing distance for most individuals),<br />
magnification is given by the relationship<br />
magnification =<br />
Thus, a 25-mm focal length positive lens would be a 10! magnifier.<br />
DIOPTERS<br />
Diopter is a term used to define the reciprocal of the focal length,<br />
which is commonly used for ophthalmic lenses. The inverse focal<br />
length of a lens expressed in diopters is<br />
diopters = 1000<br />
f<br />
250 mm<br />
f<br />
Thus, the smaller the focal length, the larger the power in diopters.<br />
DEPTH OF FIELD AND DEPTH OF FOCUS<br />
(f in mm). (1.32)<br />
(f in mm). (1.33)<br />
In an imaging system, depth of field refers to the distance in<br />
object space over which the system delivers an acceptably sharp<br />
image. The criteria for what is acceptably sharp is arbitrarily chosen<br />
by the user; depth of field increases with increasing f-number.<br />
For an imaging system, depth of focus is the range in image<br />
space over which the system delivers an acceptably sharp image. In<br />
other words, this is the amount that the image surface (such as a<br />
screen or piece of photographic film) could be moved while maintaining<br />
acceptable focus. Again, criteria for acceptability are defined<br />
arbitrarily.<br />
In nonimaging applications, such as laser focusing, depth of<br />
focus refers to the range in image space over which the focused<br />
spot diameter remains below an arbitrary limit.<br />
APPLICATION NOTE<br />
Technical Reference<br />
For further reading about the definitions and<br />
formulas presented here, refer to the following<br />
publications:<br />
Rudolph Kingslake, Lens Design Fundamentals<br />
(Academic Press)<br />
Rudolph Kingslake, <strong>Optical</strong> System Design<br />
(Academic Press)<br />
Warren Smith, Modern <strong>Optical</strong> Engineering<br />
(McGraw Hill).<br />
If you need help with the use of definitions and<br />
formulas presented in this catalog, our applications<br />
engineers will be pleased to assist you.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.31
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Paraxial Lens Formulas<br />
PARAXIAL FORMULAS FOR LENSES IN AIR<br />
The following formulas are based on the behavior of paraxial<br />
rays, which are always very close and nearly parallel to the optical<br />
axis. In this region, lens surfaces are always very nearly normal to<br />
the optical axis, and hence all angles of incidence and refraction<br />
are small. As a result, the sines of the angles of incidence and<br />
refraction are small (as used in Snell’s law) and can be approximated<br />
by the angles themselves (measured in radians).<br />
The paraxial formulas do not include effects of spherical<br />
aberration experienced by a marginal ray — a ray passing through<br />
the lens near its edge or margin. All effective focal length values (f)<br />
tabulated in this catalog are paraxial values which correspond to the<br />
paraxial formulas.<br />
The following paraxial formulas are valid for both thick and<br />
thin lenses unless otherwise noted. The refractive index of the lens<br />
glass, n, is the ratio of the speed of light in vacuum to the speed of<br />
light in the lens glass. All other variables are defined in figure 1.33.<br />
Focal Length<br />
1<br />
⎛ 1 1 ⎞ (n 41)<br />
= (n 41)<br />
4 +<br />
f<br />
⎜<br />
⎝ r r<br />
⎟<br />
⎠ n<br />
t c<br />
rr<br />
where n is the refractive index, t c is the center thickness, and the<br />
sign convention previously given for the radii r 1 and r 2 applies. For<br />
thin lenses, t c ≅ 0, and for plano lenses either r 1 or r 2 is infinite. In<br />
either case the second term of the above equation vanishes, and we<br />
are left with the familiar Lens Maker’s formula<br />
1<br />
f<br />
⎛<br />
= (n 41)<br />
⎜ 4<br />
⎝<br />
2<br />
1 1 ⎞<br />
⎟<br />
r1 r2⎠<br />
(1.35)<br />
(1.34)<br />
1 2<br />
1 2<br />
s>0<br />
Surface Sagitta and Radius of Curvature<br />
(refer to figure 1.34)<br />
2 2 ⎛ d ⎞<br />
r = (r 4s) + ⎜ ⎟<br />
⎝ 2⎠<br />
2 ⎛ d ⎞<br />
s = r 4 r 4⎜<br />
⎟<br />
⎝ 2⎠<br />
r<br />
1<br />
=<br />
f<br />
2<br />
= s 2 + d .<br />
8s<br />
2<br />
2<br />
> 0<br />
An often useful approximation is to neglect s/2.<br />
Symmetric Lens Radii (r 2 = 5r 1 )<br />
With center thickness constrained,<br />
⎡<br />
⎤<br />
ft<br />
2 c<br />
r 1 = (n 4 1)<br />
⎢ ⎛ ⎞<br />
f ± f 4<br />
⎥<br />
⎢ ⎜<br />
n<br />
⎟<br />
⎝ ⎠<br />
⎥<br />
⎣⎢<br />
⎦⎥<br />
⎡<br />
⎤<br />
tc<br />
= (n 4 1) f<br />
⎢ ⎛ ⎞<br />
1 + 14<br />
⎥<br />
⎢ ⎜<br />
⎝ nf<br />
⎟<br />
⎠<br />
⎥<br />
⎣⎢<br />
⎦⎥<br />
2 (n41)<br />
(n41)<br />
4<br />
r nr<br />
1<br />
2<br />
⎡ ⎧⎪<br />
⎛ f ⎞ ⎫⎪<br />
⎤<br />
⎢t + 2r ⎨14<br />
cos ⎜arcsin ⎝ 2r<br />
⎟ ⎬⎥<br />
⎩⎪<br />
1<br />
⎣<br />
⎢<br />
⎠ ⎭⎪ ⎦<br />
⎥<br />
(1.40)<br />
2 c 1<br />
1<br />
(1.36)<br />
(1.37)<br />
(1.38)<br />
(1.39)<br />
where, in the first form, the + sign is chosen for the square root if f is<br />
positive, but the 4 sign must be used if f is negative. In the second<br />
form, the + sign must be used regardless of the sign of f. With edge<br />
thickness constrained, the equation for r 1 becomes transcendental:<br />
where Ω is the lens diameter. This equation can be solved by numerical<br />
methods.<br />
Plano Lens Radius<br />
Since r 2 is infinite,<br />
r>0<br />
d<br />
2<br />
r 1 = (n 4 1) f.<br />
Principal-Point Locations (signed distances from vertices)<br />
(1.41)<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
(r4s)<br />
Figure 1.34 Surface sagitta and radius of curvature<br />
A H ′′ =<br />
2<br />
AH =<br />
1<br />
4rt<br />
2 c<br />
n (r 4 r ) + t (n 4 1)<br />
2 1 c<br />
4rt<br />
1 c<br />
n (r 4 r ) + t (n 4 1)<br />
2 1 c<br />
where the above sign convention applies.<br />
(1.42)<br />
(1.43)<br />
1.32 1 Visit Us OnLine! www.mellesgriot.com
For symmetric lenses (r 2 = 4r 1 ),<br />
AH = 4A H′′<br />
1 2<br />
AH = 0<br />
1<br />
=<br />
and<br />
t c<br />
A2H ′′ = 4 .<br />
n<br />
rt 1 c<br />
.<br />
2nr 4 t (n 4 1)<br />
1 c<br />
⎧<br />
⎪ f ⎡1<br />
(n 1)<br />
HH ′′ = t c ⎨1<br />
4 ⎢ 4 4<br />
n f n<br />
⎣⎢<br />
⎩⎪<br />
⎛ 1 ⎞<br />
HH ′′ = t c ⎜1<br />
4 ⎟ .<br />
⎝ n ⎠<br />
Q = 2 p(1 4cos v)<br />
⎛ v ⎞<br />
2<br />
= 4 p sin ⎜<br />
⎝ 2<br />
⎟<br />
⎠<br />
⎛ f ⎞<br />
v = arctan ⎜<br />
⎝ 2 s′′<br />
⎟<br />
⎠<br />
2<br />
t ⎤⎫<br />
c ⎪<br />
⎥⎬<br />
rr 1 2⎦⎥<br />
⎭⎪<br />
(1.44)<br />
If either r 1 or r 2 is infinite, l’Hôpital’s rule from calculus must be used.<br />
Thus, referring to page 1.27, for plano-convex lenses in the correct<br />
orientation,<br />
(1.45)<br />
For flat plates, by letting r 1 →∞ in a symmetric lens, we obtain<br />
A 1 H = A 2 H″ = t c /2n. These results are useful in connection with<br />
the following paraxial lens combination formulas.<br />
Hiatus or Interstitium (principal-point separation)<br />
(1.46)<br />
which, in the thin-lens approximation (exact for plano lenses),<br />
becomes<br />
(1.47)<br />
Solid Angle<br />
The solid angle subtended by a lens, for an observer situated at an<br />
on-axis image point, is<br />
where this result is in steradians, and where<br />
(1.48)<br />
is the apparent angular radius of the lens clear aperture. For an<br />
observer at an on-axis object point, use s instead of s″. To convert<br />
from steradians to the more intuitive sphere units, simply divide<br />
Q by 4p. If the Abbé sine condition is known to apply, ß may<br />
be calculated using the arc sine function instead of the arc<br />
tangent.<br />
Back Focal Length<br />
and<br />
f = f" + AH′′<br />
b 2<br />
= f " 4<br />
t c<br />
f b = f " 4<br />
n .<br />
f = f 4 AH<br />
f 1<br />
= f +<br />
rt 2 c<br />
n(r 4 r ) + t (n 4 1)<br />
2 1 c<br />
rt 1 c<br />
n(r 4 r ) + t (n 4 1)<br />
2 1 c<br />
m = s ′′<br />
s<br />
f<br />
=<br />
s 4 f<br />
= s ′′ 4 f .<br />
f<br />
PARAXIAL FORMULAS FOR<br />
LENSES IN ARBITRARY MEDIA<br />
(1.49)<br />
where the sign convention presented above applies to A 2 H″ and to<br />
the radii. If r 2 is infinite, l’Hôpital’s rule from calculus must be used,<br />
whereby<br />
Front Focal Length<br />
(1.50)<br />
(1.50)<br />
(1.51)<br />
where the sign convention presented above applies to A 1 H and to<br />
the radii. If r 1 is infinite, l’Hôpital’s rule from calculus must be used,<br />
whereby<br />
t c<br />
f f = f 4<br />
n .<br />
(1.52)<br />
Edge-to-Focus Distances<br />
For positive lenses,<br />
A = f + s<br />
f 1<br />
B = f + s<br />
b 2<br />
(1.53)<br />
(1.54)<br />
where s 1 and s 2 are the sagittas of the first and second surfaces.<br />
Bevel is neglected.<br />
Magnification or Conjugate Ratio<br />
(1.55)<br />
These formulas allow for the possibility of distinct and completely<br />
arbitrary refractive indices for the object space medium (refractive<br />
index n′), lens (refractive index n″), and image space medium (refractive<br />
index n). In this situation, the effective focal length assumes two<br />
distinct values, namely f in object space and f″ in image space. It is<br />
also necessary to distinguish the principal points from the nodal<br />
points. The lens serves both as a lens and as a window separating<br />
the object space and image space media.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.33
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
The situation of a lens immersed in a homogenous fluid (figure<br />
1.35) is included as a special case (n = n″). This case is of<br />
considerable practical importance. The two values f and f″ are again<br />
equal, so that the lens-combination formulas are applicable to<br />
systems immersed in a common fluid. The general case (two different<br />
fluids) is more difficult, and it must be approached by ray tracing on<br />
a surface-by-surface basis.<br />
LENS CONSTANT (k)<br />
This number appears frequently in the following formulas. It is<br />
an explicit function of the complete lens prescription (both radii,<br />
t c and n′ ) and both media indices (n and n″). This dependence is<br />
implicit anywhere that k appears.<br />
k = n ′ 4 n + n ′′ 4 n ′ t c(n′ 4n)(n′′ 4 n ′)<br />
4 . (1.56)<br />
r r<br />
nrr ′<br />
f = n k<br />
n<br />
s + n ′′<br />
s′′<br />
1 2<br />
Effective Focal Lengths<br />
f ′′ = n ′′<br />
k .<br />
Lens Formula (Gaussian form)<br />
= k.<br />
Lens Formula (Newtonian form)<br />
xx ′′ = ff ′′ = nn ′′<br />
k 2<br />
where x = s4f and x″ = s″4f ″.<br />
index n = 1 (air or vacuum)<br />
f f<br />
1 2<br />
f<br />
(1.57)<br />
(1.58)<br />
(1.59)<br />
Principal-Point Locations<br />
nt c<br />
⎛ n ′′ 4n′<br />
⎞<br />
AH 1 = ⎜ ⎟<br />
(1.60)<br />
k ⎝ nr ′ 2 ⎠<br />
4n′′ tc<br />
⎛ n ′ 4 n⎞<br />
A2H ′′ = ⎜ ⎟ . (1.61)<br />
k ⎝ nr ′ ⎠<br />
Object-to-First-Principal-Point Distance<br />
s =<br />
s ′′ =<br />
ns′′<br />
.<br />
ks ′′ 4 n′′<br />
Second Principal-Point-to-Image Distance<br />
Magnification<br />
m = ns ′′<br />
.<br />
n′′<br />
s<br />
index n"= 1.333 (water)<br />
f″<br />
f b<br />
1<br />
n′′<br />
s<br />
. (1.63)<br />
ks 4 n<br />
Lens Maker’s Formula<br />
n<br />
f<br />
= n ′′<br />
f ′′<br />
AN 1 = AH+HN<br />
1<br />
A N ′′ = A H ′′ +H′′ N ′′.<br />
2 2<br />
= k.<br />
Nodal-Point Locations<br />
(1.62)<br />
(1.64)<br />
(1.65)<br />
(1.66)<br />
(1.67)<br />
A 1 A 2<br />
F<br />
H H″ N N″<br />
F″<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 1.35<br />
Symmetric lens with disparate object and image space indices<br />
index n′ = 1.51872 (BK7)<br />
1.34 1 Visit Us OnLine! www.mellesgriot.com
Separation of Nodal Point<br />
from Corresponding Principal Point<br />
HN = H″N″ = (n″4n)/k, positive for N to right of H<br />
and N″ to right of H″.<br />
Back Focal Length<br />
f b = f ′′ + AH. 2 ′′ (see eq. 1.49)<br />
Front Focal Length<br />
f f = f 4 A1H.<br />
(see eq. 1.51)<br />
Focal Ratios<br />
The focal ratios are f/f and f ″/f, where f is the diameter of the<br />
clear aperture of the lens.<br />
Numerical Apertures<br />
n sin v<br />
⎛ f ⎞<br />
where v = arcsin ⎜<br />
⎝ 2s<br />
⎟<br />
⎠<br />
and<br />
n′′<br />
sin v"<br />
⎛ f ⎞<br />
where v" = arcsin ⎜<br />
⎝ 2s ′′<br />
⎟<br />
⎠<br />
.<br />
Solid Angles (in steradians)<br />
Q = 2 p(1 4 cos v)<br />
⎛ v ⎞<br />
2<br />
= 4p<br />
sin ⎜<br />
⎝ 2<br />
⎟<br />
⎠<br />
⎛ f ⎞<br />
where v = arctan ⎜<br />
⎝ 2s<br />
⎟<br />
⎠<br />
(see eq.1.48)<br />
Q ″ = 2p 2 p<br />
(1 4 cos v″)<br />
′′)<br />
⎛ v ⎞<br />
2<br />
= 4p<br />
sin ⎜<br />
⎝ 2<br />
⎟<br />
(1.68)<br />
⎠<br />
⎛ f ⎞<br />
where v′′<br />
= arctan ⎜<br />
⎝ 2s ′′<br />
⎟<br />
⎠<br />
.<br />
To convert from steradians to spheres, simply divide by 4p.<br />
APPLICATION NOTE<br />
For Quick Approximations<br />
Much time and effort can be saved by ignoring the<br />
differences among f, f b , and f f in these formulas<br />
(assume f = f b = f f ) by thinking of s as the lens-toobject<br />
distance, by thinking of s″ as the lens-to-image<br />
distance, and by thinking of the sum of conjugate<br />
distances s + s″ as being the object-to-image distance.<br />
This is known as the thin-lens approximation.<br />
APPLICATION NOTE<br />
Physical Significance of the Nodal Points<br />
A ray directed at the primary nodal point N of a lens<br />
appears to emerge from the secondary nodal point<br />
N″ without change of direction. Conversely, a ray<br />
directed at N″ appears to emerge from N without<br />
change of direction. At the infinite conjugate ratio,<br />
if a lens is rotated about a rotational axis orthogonal<br />
to the optical axis at the secondary nodal point<br />
(i.e., if N″ is the center of rotation), the image<br />
remains stationary during the rotation. This fact<br />
is the basis for the nodal slide method for measuring<br />
nodal-point location. The nodal points coincide with<br />
their corresponding principal points when the image<br />
space and object space refractive indices are equal (n<br />
= n″). This makes the nodal slide method the most<br />
precise method of principal-point location.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 1.35
Fundamental Optics<br />
Principal-Point Locations<br />
Figure 1.36 indicates approximately where the principal points fall<br />
in relation to the lens surfaces for various standard lens shapes. The<br />
exact positions depend on the index of refraction of the lens material,<br />
and on the lens radii, and can be found by formula. In extreme<br />
meniscus lens shapes (short radii or steep curves), it is possible that<br />
both principal points will fall outside the lens boundaries. For<br />
symmetric lenses, the principal points divide that part of the optical<br />
axis between the vertices into three approximately equal segments.<br />
For plano lenses, one principal point is at the curved vertex, and the<br />
other is approximately one-third of the way to the plane vertex.<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
F″<br />
H″ F″ H″<br />
H″<br />
F″<br />
F″<br />
H″<br />
F″<br />
F″<br />
H″<br />
H″<br />
F″<br />
F″<br />
H″<br />
H″<br />
H″<br />
F″<br />
F″<br />
H″<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 1.36<br />
Principal points of common lenses<br />
1.36 1 Visit Us OnLine! www.mellesgriot.com
Gaussian Beam Optics 2<br />
Introduction to Gaussian Beam Optics 2.2<br />
Transformation and Magnification by Simple Lenses 2.6<br />
Lens Selection 2.10<br />
2.1 1<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong>
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Introduction to Gaussian Beam Optics<br />
In most laser applications it is necessary to focus, modify, or<br />
shape the laser beam by using lenses and other optical elements. In<br />
general, laser-beam propagation can be approximated by assuming<br />
that the laser beam has an ideal Gaussian intensity profile,<br />
corresponding to the theoretical TEM 00 mode. Coherent Gaussian<br />
beams have peculiar transformation properties that require special<br />
consideration. In order to select the best optics for a particular laser<br />
application, it is important to understand the basic properties of<br />
Gaussian beams. Unfortunately, the output from real-life lasers is<br />
not truly Gaussian (although helium neon lasers and argon-ion<br />
lasers are a very close approximation). To accommodate this variance,<br />
a quality factor, M 2 (called the “M-square” factor), has been defined<br />
to describe the deviation of the laser beam from a theoretical<br />
Gaussian. For a theoretical Gaussian, M 2 =1; for a real laser beam,<br />
M 2 >1. Helium neon lasers typically have an M 2 factor that is less<br />
than 1.1. For ion lasers, the M 2 factor is typically between 1.1 and<br />
1.3. Collimated TEM 00 diode laser beams usually have an M 2 factor<br />
ranging from 1.1 to 1.7. For high-energy multimode lasers, the M 2<br />
factor can be as high as 3 or 4. In all cases, the M 2 factor, which<br />
varies significantly, affects the characteristics of a laser beam and<br />
cannot be neglected in optical designs.<br />
In the following discussion, we will first treat the characteristics<br />
of a theoretical Gaussian beam (M 2 = 1) and then show how these<br />
characteristics change as the beam deviates from the theoretical. In<br />
all cases, a circularly symmetric wavefront is assumed, as would be<br />
the case for a helium neon laser or an argon-ion laser. Diode laser<br />
beams are asymmetric and often astigmatic, which causes their<br />
transformation to be more complex.<br />
Although in some respects component design and tolerancing<br />
for lasers are more critical than they are for conventional optical<br />
components, the designs often tend to be simpler since many of<br />
the constants associated with imaging systems are not present. For<br />
instance, laser beams are nearly always used on axis, which eliminates<br />
the need to correct asymmetric aberration. Chromatic aberrations<br />
are of no concern in single-wavelength lasers, although they are<br />
critical for some tunable and multiline laser applications. In fact, the<br />
only significant aberration in most single-wavelength applications<br />
is primary (third-order) spherical aberration.<br />
Scatter from surface defects, inclusions, dust, or damaged coatings<br />
is of greater concern in laser-based systems than in incoherent<br />
systems. Speckle content arising from surface texture and beam<br />
coherence can limit system performance.<br />
Because laser light is generated coherently, it is not subject to<br />
some of the limitations normally associated with incoherent sources.<br />
All parts of the wavefront act as if they originate from the same<br />
point, and consequently the emergent wavefront can be precisely<br />
defined. Starting out with a well-defined wavefront permits more<br />
precise focusing and control of the beam than would otherwise be<br />
possible.<br />
In order to gain an appreciation of the principles and limitations<br />
of Gaussian beam optics, it is necessary to understand the nature of<br />
the laser output beam. In TEM 00 mode, the beam emitted from a laser<br />
is a perfect plane wave with a Gaussian transverse irradiance profile<br />
as shown in figure 2.1. The Gaussian shape is truncated at some<br />
diameter either by the internal dimensions of the laser or by some<br />
limiting aperture in the optical train. To specify and discuss propagation<br />
characteristics of a laser beam, we must define its diameter<br />
in some way. The commonly adopted definition is the diameter at<br />
which the beam irradiance (intensity) has fallen to 1/e 2 (13.5%) of its<br />
peak, or axial, value.<br />
BEAM WAIST AND DIVERGENCE<br />
Diffraction causes light waves to spread transversely as they<br />
propagate, and it is therefore impossible to have a perfectly collimated<br />
beam. The spreading of a laser beam is in precise accord with the<br />
predictions of pure diffraction theory; aberration is totally insignificant<br />
in the present context. Under quite ordinary circumstances,<br />
the beam spreading can be so small it can go unnoticed. The following<br />
formulas accurately describe beam spreading, making it<br />
easy to see the capabilities and limitations of laser beams. The<br />
notation is consistent with much of the laser literature, particularly<br />
with Siegman’s excellent Introduction to Lasers and Masers<br />
(McGraw-Hill).<br />
PERCENT IRRADIANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
13.5<br />
Figure 2.1<br />
41.5w 4w 0 w 1.5w<br />
CONTOUR RADIUS<br />
Irradiance profile of a Gaussian TEM 00 mode<br />
2.2 1 Visit Us OnLine! www.mellesgriot.com
Even if a Gaussian TEM 00 laser-beam wavefront were made<br />
perfectly flat at some plane, with all elements moving in precisely<br />
parallel directions, it would quickly acquire curvature and begin<br />
spreading in accordance with<br />
⎡<br />
2<br />
⎛ 2<br />
p w ⎞ ⎤<br />
0<br />
R(z) = z<br />
⎢<br />
1 +<br />
⎥<br />
⎢ ⎜ ⎟<br />
⎝ lz<br />
⎥<br />
⎠<br />
⎣⎢<br />
⎦⎥<br />
and<br />
⎡<br />
lz<br />
w(z) = w 0 + ⎛ 2<br />
⎞ ⎤<br />
⎢1<br />
2<br />
⎝ ⎜ ⎥<br />
⎢ ⎟<br />
p w0<br />
⎠ ⎥<br />
⎣<br />
⎦<br />
12 /<br />
where z is the distance propagated from the plane where the wavefront<br />
is flat, l is the wavelength of light, w 0 is the radius of<br />
the 1/e 2 irradiance contour at the plane where the wavefront is flat, w(z)<br />
is the radius of the 1/e 2 contour after the wave has propagated a<br />
distance z, and R(z) is the wavefront radius of curvature after<br />
propagating a distance z. R(z) is infinite at z = 0, passes through<br />
a minimum at some finite z, and rises again toward infinity as<br />
z is further increased, asymptotically approaching the value of z itself.<br />
The plane z = 0 marks the location of a Gaussian waist, or a place<br />
where the wavefront is flat, and w 0 is called the beam waist radius.<br />
A waist occurs naturally at the midplane of a symmetric confocal<br />
cavity. Another waist occurs at the surface of the planar mirror<br />
of the quasi-hemispherical cavity used in many Melles Griot lasers.<br />
The irradiance distribution of the Gaussian TEM 00 beam,<br />
namely,<br />
I (r) = I e =<br />
2P<br />
w e 2<br />
p<br />
42r<br />
2 / w<br />
2 42r<br />
2 / w<br />
2<br />
0<br />
where w = w(z) and P is the total power in the beam, is the same<br />
at all cross sections of the beam. The invariance of the form of the<br />
distribution is a special consequence of the presumed Gaussian<br />
distribution at z = 0. If a uniform irradiance distribution had been<br />
presumed at z = 0, the pattern at z = ∞ would have been the familiar<br />
Airy disc pattern given by a Bessel function, while the pattern at<br />
intermediate z values would have been enormously complicated. (See<br />
Born and Wolf, Principles of Optics, 2d ed, Pergamon/ Macmillan).<br />
Simultaneously, as R(z) asymptotically approaches z for large<br />
z, w(z) asymptotically approaches the value<br />
w(z)<br />
lz<br />
≅<br />
p<br />
w 0<br />
where z is presumed to be much larger than pw 0 /l so that the 1/e 2<br />
irradiance contours asymptotically approach a cone of angular<br />
radius<br />
v<br />
= w(z)<br />
z<br />
l<br />
= .<br />
pw 0<br />
,<br />
(2.1)<br />
(2.2)<br />
(2.3)<br />
(2.4)<br />
(2.5)<br />
This value is the far-field angular radius of the Gaussian TEM 00<br />
beam. The vertex of the cone lies at the center of the waist (see<br />
figure 2.2).<br />
It is important to note that, for a given value of l, variations of<br />
beam diameter and divergence with distance z are functions of a<br />
single parameter. This is often chosen to be w 0 , or the beam waist<br />
radius.<br />
The direct relationship between beam waist and divergence<br />
(v ∝ 1/w 0 ) must always be considered when focusing a TEM 00 laser<br />
beam. Because of this relationship, the spectrally selective coating<br />
of the spherical output mirror of a Melles Griot laser is actually supported<br />
on the concave inner surface of a weak meniscus lens. In<br />
this paraxial, high f-number configuration, the lens introduces no<br />
significant aberration. A new beam waist, larger than the intracavity<br />
beam waist, is formed by this lens near its output pupil. The<br />
transformed beam has greatly reduced divergence, which is<br />
advantageous for most applications. Note that it is the 1/e 2 beam<br />
diameter of this extracavity waist that is published in this catalog.<br />
As an example to illustrate the relationship between beam waist<br />
and divergence, let us consider the real case of a Melles Griot red<br />
5-mW HeNe laser, 05 LHR 151, with a specified beam diameter of<br />
0.8 mm (i.e., w 0 = 0.4 mm). In the far-field region,<br />
l<br />
v =<br />
pw = 632.8 ×<br />
( p)<br />
(0.4)<br />
Using the asymptotic approximation, at a distance of z = 100 m,<br />
w<br />
w 0<br />
w 0<br />
0<br />
w(z) = zv<br />
5 4<br />
= (10 )( 5.04 × 10<br />
4 )<br />
= 50.4 mm<br />
1<br />
e 2<br />
6<br />
10 5 54<br />
irradiance surface<br />
v<br />
= 5.04 × 10 rad.<br />
which is approximately 126 times larger than w 0 .<br />
asymptotic cone<br />
z<br />
w 0<br />
Figure 2.2 Growth in 1/e 2 contour radius with distance<br />
propagated away from Gaussian waist<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 2.3
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Suppose instead that we decide to reduce the divergence<br />
by directing the laser into a beam expander (reversed telescope)<br />
of angular magnification m = 10, such as Melles Griot model<br />
09 LBM 013 (figure 2.3). Consider the case in which the expander<br />
is focused to form a waist of radius w 0 = 4.0 mm at the expander<br />
output lens. Since v ∝ 1/w 0 , by definition, v is reduced by a factor<br />
of 10; therefore, for z = 100 m,<br />
For the expanded beam, the ratio w(z)/w 0 is only a factor of 12.6<br />
for a distance of 100 m, but it is a factor of 126 for the same distance<br />
when the laser is used alone.<br />
OPTIMUM COLLIMATION<br />
Typically, one has a fixed value for w 0 and uses the previously given<br />
expression to calculate w(z) for an input value of z. However, one can<br />
also utilize this equation to see how final beam radius varies with starting<br />
beam radius at a fixed distance, z. Figure 2.4 shows the Gaussian<br />
beam propagation equation plotted as a function of w 0 , with the<br />
particular values of l = 632.8 nm and z = 100 m.<br />
The beam radius at 100 m reaches a minimum value for a starting<br />
beam radius of about 4.5 mm. Therefore, if we wanted to achieve<br />
the best combination of minimum beam diameter and minimum<br />
beam spread (or best collimation) over a distance of 100 m, our<br />
optimum starting beam radius would be 4.5 mm. Any other starting<br />
value would result in a larger beam at z = 100 m.<br />
We can find the general expression for the optimum starting<br />
beam radius for a given distance, z. Doing so yields<br />
w (optimum) =<br />
0<br />
5 5 4<br />
w(z) = (10 )( 504 . × 10 )<br />
= 5.04 mm.<br />
10<br />
1/2<br />
⎛ lz<br />
⎞<br />
⎜<br />
⎝ p<br />
⎟ . (2.6)<br />
⎠<br />
Using this optimum value of w 0 will provide the best combination<br />
of minimum starting beam diameter and minimum beam<br />
spread (ratio of w(z)/w 0 ) over the distance z. The previous example<br />
of z = 100 and l=632.8 nm gives w 0 (optimum) = 4.48 mm, shown<br />
graphically in figure 2.4. If we put this value for w 0 (optimum) back<br />
into the expression for w(z), w(z) = √}}2 w 0 . Thus, for this example,<br />
w(100) = √}}2 (4.48) = 6.3 mm.<br />
By turning this previous equation around, we can define a<br />
distance, called the Rayleigh range (z R ), over which the beam radius<br />
spreads by a factor of √}}2 as<br />
z R =<br />
with<br />
If we use beam-expanding optics (such as the 09 LBC, 09 LBX,<br />
09 LBZ, or 09 LCM series), which allow us to adjust the position<br />
of the beam waist, we can actually double the distance over which<br />
beam divergence is minimized. Figure 2.5 illustrates this situation,<br />
in which the beam starts off at a value of w(z R ) = (2lz/p) 1/2 , goes<br />
through a minimum value of w 0 = w(z R )/√}}2 , and then returns to<br />
w(z R ). By focusing the beam-expanding optics to place the beam<br />
waist at the midpoint, we can restrict beam spread to a factor of √}}2<br />
over a distance of 2z R , as opposed to just z R .<br />
This result can now be used in the problem of finding the starting<br />
beam radius that yields the minimum beam diameter and beam<br />
spread over 100 m. Using 2z R = 100, or z R = 50, and l = 632.8 nm,<br />
we get a value of w(z R ) = (2lz/p) 1/2 = 4.5 mm, and w 0 = 3.2 mm.<br />
Thus, the optimum starting beam radius is the same as previously<br />
calculated. However, by focusing the expander we achieve a final<br />
beam radius that is no larger than our starting beam radius, while<br />
still maintaining the √}}2 factor in overall variation.<br />
Alternately, if we started off with a beam radius of 6.3 mm<br />
(√}}2w 0 ), we could focus the expander to provide a beam waist of<br />
w 0 = 4.5 mm at 100 m, and a final beam radius of 6.3 mm at 200 m.<br />
FINAL BEAM RADIUS (mm)<br />
100<br />
80<br />
60<br />
40<br />
20<br />
pw<br />
l<br />
2<br />
0<br />
w(z ) = 2w<br />
.<br />
R 0<br />
0<br />
0 1 2 3 4 5 6 7 8 9 10<br />
STARTING BEAM RADIUS w 0 (mm)<br />
(2.7)<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 2.3<br />
telescope)<br />
Laser beam expander 09 LBM 013 (reversed<br />
Figure 2.4 Beam radius at 100 m as a function of starting<br />
beam radius for a HeNe laser at 632.8 nm<br />
2.4 1 Visit Us OnLine! www.mellesgriot.com
eam expander<br />
INCORPORATING M 2 INTO THE BASIC EQUATIONS<br />
The following discussion is taken from the analysis by Sun [Haiyin<br />
Sun, “Thin Lens Equation for a Real Laser Beam with Weak Lens<br />
Aperture Truncation,” Opt. Eng. 37, no. 11 (November 1998)]. From<br />
equation 2.5 we see that, for a theoretical Gaussian beam, the smallest<br />
possible value of the radius-divergence product is<br />
w 0 v = l/p.<br />
For a real laser beam, we have<br />
w 0M v M = M 2 l/p >l/p<br />
where w 0M and v M are the 1/e 2 intensity waist radius and the farfield<br />
half-divergent angle of the real laser beam, respectively, and<br />
M 2 factors into equations 2.1 and 2.2 as follows:<br />
w M (z) = w 0M [1+(zlM 2 /pw 0M 2 ) 2 ] 1/2<br />
R M (z) = z[1+(pw 0M 2 /zlM 2 ) 2 ]<br />
where w M and R M are the 1/e 2 intensity radius of the beam and the<br />
beam wavefront radius at z, respectively.<br />
The definition for the Rayleigh range (equation 2.7) remains<br />
the same for a real laser beam and becomes<br />
z R = pw 0R 2 /l.<br />
z R<br />
Together, equations 2.9, 2.10, and 2.11 form a complete set to<br />
denote the input of a real laser beam into a thin lens.<br />
w 0<br />
w(–z R ) = √2w 0<br />
w(z R ) = √2w 0<br />
Figure 2.5 Focusing a beam expander to minimize beam<br />
radius and spread over a specified distance<br />
z R<br />
(2.8)<br />
(2.9)<br />
(2.10)<br />
(2.11)<br />
LASERS AND LASER SYSTEMS<br />
Melles Griot manufactures many types of lasers and<br />
laser systems for laboratory and OEM applications.<br />
These, along with a wide variety of laser accessories, are<br />
found in Chapter 41 through 47. Laser types include<br />
helium neon (HeNe) and helium cadmium (HeCd) lasers;<br />
argon, krypton, and mixed gas (argon/krypton) ion<br />
lasers; diode lasers, and diode-pumped solid-state<br />
(DPSS) lasers.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 2.5
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Transformation and Magnification by Simple Lenses<br />
It is already clear from the previous discussion that Gaussian<br />
beams transform in an unorthodox manner. Siegman uses matrix<br />
transformations to treat the general problem of Gaussian beam<br />
propagation with lenses and mirrors. A less rigorous, but in many<br />
ways more insightful, approach to this problem has been developed<br />
by Self [S.A. Self, “Focusing of Spherical Gaussian Beams,” Appl.<br />
Opt. 22, no. 5 (March 1983): 658]. Self shows a method to model<br />
transformations of a laser beam through simple optics, under<br />
paraxial conditions, by calculating the Rayleigh range and beam<br />
waist location following each individual optical element. These<br />
parameters are calculated using a formula analogous to the<br />
well-known standard lens formula. Melles Griot engineers have<br />
found this method to be particularly useful. The main points are as<br />
follows.<br />
The standard lens equation can be written in dimensionless<br />
form:<br />
1<br />
s/f<br />
+<br />
1<br />
s ″/f<br />
1<br />
s + z s f) + 1<br />
= 1 2<br />
s f<br />
R /( 4 ″<br />
or, in dimensionless form,<br />
1<br />
(s/f) + (z /f) /(s/f 1) + 1<br />
= 1.<br />
2<br />
4 (s″/f)<br />
R<br />
= 1.<br />
For Gaussian beams, Self has derived an analogous formula by<br />
assuming that the waist of the input beam represents the object,<br />
and the waist of the output beam represents the image. The formula<br />
is expressed in terms of the Rayleigh range of the input beam.<br />
In the regular form,<br />
In the far-field limit as z R → 0, this reduces to the geometric<br />
optics equation. A plot of (s/f) versus (s″/f) for various values of<br />
(z R /f) is shown in figure 2.6. There are three distinct regions of<br />
interest. For a positive thin lens, these correspond to real object<br />
and real image, real object and virtual image, and virtual object<br />
and real image.<br />
The main differences between Gaussian beam optics and<br />
geometric optics, highlighted in such a plot, can be summarized as<br />
follows:<br />
$ There is a maximum and minimum image distance for<br />
Gaussian beams.<br />
(2.12)<br />
(2.13)<br />
(2.14)<br />
$ The maximum image distance occurs at s = f + z R , rather than<br />
at s = f.<br />
$ There is a common point in the Gaussian beam expression<br />
at s/f = s″/f =1. For a simple positive lens, this is the point at<br />
which the incident beam has a waist at the front focus and the<br />
emerging beam has a waist at the rear focus.<br />
IMAGE DISTANCE (s"/f)<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0<br />
41<br />
42<br />
43<br />
44<br />
45<br />
parameter<br />
44<br />
43<br />
z<br />
(<br />
R<br />
f )<br />
42<br />
0<br />
0.25<br />
0.50<br />
1 2<br />
41 0 1 2 3 4 5<br />
OBJECT DISTANCE<br />
(s/f)<br />
Figure 2.6 Plot of the lens formula for Gaussian beams,<br />
with normalized Rayleigh range of the input beam as<br />
the parameter<br />
$ A lens appears to have a shorter focal length as z R /f increases<br />
from zero (i.e., there is a Gaussian focal shift).<br />
Self recommends calculating z R , w 0 , and the position of w 0 for<br />
each optical element in the system in turn so that the overall transformation<br />
of the beam can be calculated. To carry this out, it is<br />
also necessary to consider magnification: w 0 ″/w 0 . The magnification<br />
is given by<br />
m = w 0″<br />
w = 1<br />
.<br />
0 ⎧ 2 2<br />
⎨[ 14<br />
(s/f)]<br />
+(z R/f)<br />
⎫<br />
⎬<br />
⎩<br />
⎭<br />
The Rayleigh range of the output beam depends on m 2 , as can<br />
be seen from the previous example, and is given by<br />
z<br />
″ = m z .<br />
R<br />
2<br />
R<br />
All the above formulas are written in terms of the Rayleigh range<br />
of the input beam. Unlike the geometric case, the formulas are not<br />
symmetric with respect to input and output beam parameters. For<br />
back tracing beams, it is useful to know the Gaussian beam formula<br />
in terms of the Rayleigh range of the output beam:<br />
1<br />
s + 1<br />
2<br />
s ″ + z ″ /(s″<br />
4 f )<br />
R<br />
= 1 f .<br />
(2.15)<br />
(2.16)<br />
(2.17)<br />
2.6 1 Visit Us OnLine! www.mellesgriot.com
M 2 AND THE LENS EQUATION<br />
For real-world beams, the lens equation can be modified to<br />
incorporate M 2 . Equation 2.12 becomes<br />
1/[s+(z R /M 2 ) 2 /(s-f)]+1/2″ = 1/f,<br />
and equation 2.14 transforms to<br />
1/[(s/f)+(z R /M 2 f) 2 /(s/f-1)]+1/(s″/f) = 1.<br />
BEAM CONCENTRATION<br />
The spot size and focal position of a Gaussian beam can be<br />
determined from the previous equations. Two cases of particular<br />
interest occur when s = 0 (the input waist is at the first principal<br />
surface of the lens system) and s = f (the input waist is at the front<br />
focal point of the optical system). For s = 0, we get<br />
and<br />
s ″ =<br />
w =<br />
f<br />
1 + ( lf/ pw<br />
)<br />
2 2<br />
0<br />
lf/ pw0<br />
2<br />
[ 1 + ( lf/ pw0<br />
) ]<br />
/<br />
2 12<br />
.<br />
For the case of s = f, the equations for image distance and waist<br />
size reduce to the following:<br />
and<br />
s ″ = f<br />
w = lf/ p w 0<br />
.<br />
Substituting typical values into these equations yields nearly<br />
identical results, and for most applications, the simpler, second set<br />
of equations can be used.<br />
In many applications, a primary aim is to focus the laser to a very<br />
small spot, as shown in figure 2.7, by using either a single lens or a<br />
combination of several lenses. Melles Griot has designed a series of<br />
single lenses optimized for this specific purpose. For example, by<br />
using a 05 LHR 151 laser and a focusing singlet, 01 LFS 033, the<br />
formula should be modified as follows:<br />
46<br />
4lf<br />
w(z)<br />
3 w = 4(632.8 × 10 )( 7)<br />
≅<br />
p<br />
( 3)( 0.<br />
4p)<br />
43<br />
= 4.70 × 10 mm<br />
= 4.7 mm.<br />
(2.18)<br />
(2.19)<br />
(2.20)<br />
(2.21)<br />
The factor 4/3 arises because of the careful balance of spherical<br />
aberration and diffraction designed into the singlet. The ratio f/w<br />
is proportional to lens f-number, but is not equal to it.<br />
If a particularly small spot is desired, there is an advantage to<br />
using a well-corrected high-numerical-aperture microscope objective<br />
(see Chapter 29, Microscope Components, Spatial Filters and<br />
Apertures) to concentrate the laser beam. The principal advantage<br />
of the microscope objective over a simple lens is the diminished<br />
level of spherical aberration. Although microscope objectives are<br />
often used for this purpose, they are never designed for use at the<br />
infinite conjugate ratio. Suitably optimized lens systems, which<br />
Melles Griot can design and build on special request, are more<br />
effective in beam-concentration tasks.<br />
DEPTH OF FOCUS<br />
Depth of focus (±D z), that is, the range in image space over<br />
which the focused spot diameter remains below an arbitrary limit,<br />
can be derived from the formula<br />
⎡<br />
2<br />
⎛ lz<br />
⎞ ⎤<br />
w(z) = w ⎢<br />
0 1 + ⎥<br />
⎢ ⎜ 2 ⎟<br />
⎝ pw0<br />
⎠ ⎥<br />
⎣<br />
⎦<br />
Dz<br />
2<br />
0.32pw 0<br />
≈ ± .<br />
l<br />
12 /<br />
The first step in performing a depth-of-focus calculation is to set<br />
the allowable degree of spot size variation. If we choose a typical<br />
value of 5%, or w(z) = 1.05w 0 , and solve for z = D z, the result is<br />
By applying this result to the combination of the 05 LHR 151<br />
laser and laser-line focusing singlet 01 LFS 033, we find<br />
Dz = 0.32 p( 470 . × 10 )<br />
±<br />
47<br />
6328 × 10<br />
= ± 35.1 mm.<br />
.<br />
43<br />
2<br />
(2.22)<br />
Since the depth of focus is proportional to the square of focal<br />
spot size, and focal spot size is directly related to f-number, the<br />
depth of focus is proportional to the square of the f-number of the<br />
focusing system.<br />
2w 0<br />
w<br />
1<br />
D beam<br />
e 2<br />
Figure 2.7 Concentration of a laser beam by a laser-line<br />
focusing singlet<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 2.7
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
TRUNCATION<br />
In a diffraction-limited lens, the diameter of the image spot is<br />
d = K<br />
INTENSITY<br />
1.0<br />
.9<br />
.8<br />
.7<br />
.6<br />
.5<br />
.4<br />
.3<br />
.2<br />
.1<br />
Figure 2.8<br />
plane<br />
INTENSITY<br />
1.0<br />
.9<br />
.8<br />
.7<br />
.6<br />
.5<br />
.4<br />
.3<br />
.2<br />
.1<br />
Figure 2.9<br />
plane<br />
× l ×<br />
f/#<br />
50%<br />
intensity<br />
13.5%<br />
intensity<br />
2.44 l (f-number)<br />
Airy disc intensity distribution at the image<br />
1.83 l (f-number)<br />
50%<br />
intensity<br />
13.5%<br />
intensity<br />
Gaussian intensity distribution at the image<br />
(2.23)<br />
where K is a constant dependent on truncation ratio and pupil<br />
illumination, l is the wavelength of light, and f/# is the speed of the<br />
lens at truncation. The intensity profile of the spot is strongly dependent<br />
on the intensity profile of the radiation filling the entrance<br />
pupil of the lens. For uniform pupil illumination, the image spot takes<br />
on an Airy disc intensity profile as shown in figure 2.8. If the pupil<br />
illumination is Gaussian in profile, an image spot of Gaussian<br />
profile results as shown in figure 2.9. When the pupil illumination<br />
is between these two extremes, a hybrid intensity profile results.<br />
In the case of the Airy disc, the intensity falls to zero at the<br />
point d zero = 2.44 ! l ! f/#, defining the diameter of the spot (see<br />
figure 2.8). When the pupil illumination is not uniform, the image<br />
spot intensity never falls to zero making it necessary to define the<br />
diameter at some other point. This is commonly done for two<br />
points:<br />
d FWHM = 50% intensity point<br />
and<br />
d 2 = 13.5% intensity point.<br />
1/e<br />
It is helpful to introduce the truncation ratio<br />
T = D D<br />
The k function, plotted in figure 2.10, permits calculation of<br />
on-axis spot diameter for any beam truncation ratio.<br />
The optimal choice for truncation ratio depends on the relative<br />
importance of spot size, peak spot intensity, and total power in the<br />
spot as demonstrated in the table below. The total power loss in<br />
the spot can be calculated by using<br />
42(D /D<br />
P = e t b ) 2 (2.27)<br />
L<br />
b<br />
t<br />
K = 1.<br />
029 +<br />
0.7125<br />
(T 4 0.2161)<br />
(2.24)<br />
where D b is the Gaussian beam diameter measured at the 1/e 2<br />
intensity point, and D t is the limiting aperture diameter of the lens.<br />
If T = 2, which approximates uniform illumination, the image spot<br />
intensity profile approaches that of the classic Airy disc. When<br />
T = 1, the Gaussian profile is truncated at the 1/e 2 diameter, and the<br />
spot profile is clearly a hybrid between an Airy pattern and a<br />
Gaussian distribution. When T = 0.5, which approximates the case<br />
for an untruncated Gaussian input beam, the spot intensity profile<br />
approaches a Gaussian distribution.<br />
Calculation of spot diameter for these or other truncation ratios<br />
requires that K be evaluated. This is done by using the formulas<br />
and<br />
0.6445<br />
4<br />
(T 4 0.2161)<br />
(2.25)<br />
FWHM 2.179 2.221<br />
K = 1.6449 +<br />
2<br />
1/e<br />
4<br />
0.6460<br />
0.<br />
5320<br />
4<br />
(T 0.2816) (T 4 0.2816)<br />
1.821 1.891<br />
for a truncated Gaussian beam. A good compromise between power<br />
loss and spot size is often a truncation ratio of one. When T = 2<br />
(approximately uniform illumination), fractional power loss is 60%.<br />
When T = 1, d 1/e<br />
2 is just 8.0% larger than when T = 2, while fractional<br />
power loss is down to 13.5%. Because of this large savings in power<br />
with relatively little growth in the spot diameter, truncation ratios<br />
of 0.7 to 1.0 are typically used. Ratios as low as 0.5 might be<br />
.<br />
(2.26)<br />
2.8 1 Visit Us OnLine! www.mellesgriot.com
employed when laser power must be conserved. However, this low<br />
value often wastes too much of the available clear aperture of the<br />
lens.<br />
The mathematics of the effects of truncation on a real-world<br />
laser beam are beyond the scope of this chapter. Suffice it to say that<br />
truncation, in general, increases the M 2 factor of the beam. For an<br />
in-depth treatment of this problem, please refer to the<br />
aforementioned paper by Haiyin Sun as well as “Changes in<br />
Characteristics of a Gaussian Beam Weakly Diffracted by a Circular<br />
Aperture” by P. Belland and J. Crenn, App. Opt. 21 (1982).<br />
Spot Diameters and Fractional Power Loss<br />
for Three Values of Truncation<br />
Truncation Ratio d FWHM d 1/e<br />
2 d zero P L (%)<br />
Infinity<br />
2.0<br />
1.0<br />
0.5<br />
SPATIAL FILTERING<br />
1.03<br />
1.05<br />
1.13<br />
1.54<br />
1.64<br />
1.69<br />
1.83<br />
2.51<br />
2.44<br />
—<br />
—<br />
—<br />
100<br />
60<br />
13.5<br />
0.03<br />
Laser light scattered from dust particles residing on optical<br />
surfaces may produce interference patterns resembling holographic<br />
zone planes. Such patterns can cause difficulties in interferometric<br />
and holographic applications where they form a highly detailed,<br />
contrasting, and confusing background that interferes with desired<br />
information. Spatial filtering is a simple way of suppressing this<br />
interference and maintaining a very smooth beam irradiance distribution.<br />
The scattered light propagates in different directions from<br />
the laser light and hence is spatially separated at a lens focal plane.<br />
By centering a small aperture around the focal spot of the direct<br />
beam, it is possible to block scattered light while allowing the direct<br />
K FACTOR<br />
3.0<br />
2.5<br />
2.0<br />
1.5<br />
1.0<br />
0.5<br />
0<br />
Figure 2.10<br />
spot measured at 13.5% intensity level<br />
spot measured at 50% intensity level<br />
spot diameter = K ! l ! f-number<br />
1.0 2.0 3.0 4.0<br />
T(Db/D t )<br />
K factors as a function of truncation ratio<br />
beam to pass unscathed. The result is a cone of light that has a very<br />
smooth irradiance distribution and can be refocused to form a<br />
collimated beam that is almost equally smooth (see figure 2.11).<br />
As a compromise between ease of alignment and complete<br />
spatial filtering, it is best that the aperture diameter be about two<br />
times the 1/e 2 beam contour at the focus, or about 1.33 times the<br />
99% throughput contour diameter.<br />
Figure 2.11 Spatial filtering smoothes the irradiance<br />
distribution<br />
APPLICATION NOTE<br />
Modular and Multiaxis Spatial Filters<br />
The Melles Griot range of spatial filters includes<br />
a three-axis unit with precision micrometers<br />
(07 SFM 001) and a compact five-axis version<br />
(07 SFM 003). These devices feature an open design<br />
that provides access to the beam as it passes<br />
through the instrument. Details of these products<br />
and standard microscope objectives and mounted<br />
pinholes that work with these spatial filters are<br />
described in Chapter 29, Microscope Components,<br />
Spatial Filters, and Apertures.<br />
For those who wish to fabricate their own spatial<br />
filters, unmounted pinholes can also be found in<br />
Chapter 29, Microscope Components, Spatial Filters,<br />
and Apertures. The precision individual pinholes are<br />
for general-purpose spatial-filtering tasks. The highenergy<br />
laser precision pinholes are constructed<br />
specifically to withstand irradiation from high-energy<br />
lasers.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 2.9
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Lens Selection<br />
The most important relationships that we will use in the process tables list beam diameter, so remember to divide by 2). Assuming a<br />
of lens selection for Gaussian beam optical systems are as follows: collimated beam, we use the propagation formula to determine the<br />
spot size at 80 m:<br />
Focused spot radius<br />
⎡<br />
2<br />
l<br />
w= f ⎛<br />
3 ⎞ ⎤<br />
12 /<br />
5<br />
⎢ 0.6328 × 10 × 80,<br />
000 ⎥<br />
.<br />
(from 2.4) w (80 m) = 0.4⎢1<br />
+ ⎜<br />
⎟<br />
p w 0<br />
⎢ ⎜<br />
⎝ ( p)( 04 . ) ⎟<br />
⎥<br />
2 ⎠ ⎥<br />
⎣<br />
⎦<br />
Beam propagation<br />
= 40.3- mm beam radius<br />
or 80.6-mm beam diameter. This is just about exactly a factor of 10<br />
⎡<br />
2 12 /<br />
z<br />
w(z) = w0<br />
1+ ⎛ ⎞ ⎤<br />
⎢<br />
l<br />
larger than we wanted. We can use the formula for w<br />
⎢ 2<br />
⎝ ⎜ ⎥<br />
0 (optimum)<br />
(from 2.2)<br />
⎟<br />
pw0<br />
⎠ ⎥<br />
to determine the smallest collimated beam diameter we could<br />
⎣ ⎦<br />
achieve at a distance of 80 m:<br />
12 /<br />
⎛ lz⎞<br />
1/2<br />
3<br />
w 0 (optimum) =<br />
⎛<br />
4<br />
⎜ ⎟<br />
0.6328 × 10 × 80,000⎞<br />
⎝ p ⎠<br />
w 0 (optimum) = ⎜<br />
⎟ = 4.0 mm.<br />
⎝<br />
p<br />
and<br />
⎠<br />
2<br />
pw<br />
This tells us that if we expand the beam by a factor of 10<br />
0<br />
(from 2.7)<br />
z R = .<br />
(4.0 mm/0.4 mm), we can produce a collimated beam 8 mm in<br />
l<br />
diameter, which, if focused at the midpoint (40 m), will again be<br />
We can also utilize the equation for the approximate on-axis<br />
8 mm in diameter at a distance of 80 m. This 10# expansion could<br />
spot size caused by spherical aberration for a plano-convex lens at<br />
be accomplished most easily with one of the Melles Griot beam<br />
the infinite conjugate:<br />
expanders, such as the 09 LBX 003 or 09 LBM 013. However, if there<br />
spot diameter (3rd -order spherical aberration) = 0.067 f is a space constraint and a need to perform this task with a system<br />
(f/#) . 3<br />
This formula is for uniform illumination, not a Gaussian intensity<br />
profile. However, since it yields a larger value for spot size than actually<br />
occurs, its use will provide us with conservative lens choices.<br />
Keep in mind that this formula is for spot diameter whereas the<br />
Gaussian beam formulas are all stated in terms of spot radius.<br />
Example 1: Obtain 8-mm spot at 80 m<br />
Using the Melles Griot HeNe laser 05 LHR 151, produce a spot<br />
8 mm in diameter at a distance of 80 m (see figure 2.12).<br />
The product tables in Chapter 44, Helium Neon Lasers, gives the<br />
output beam radius for the 25 LHR 151 as 0.4 mm (the product<br />
that is no longer than 50 mm, this can be accomplished by using<br />
catalog components.<br />
Figure 2.13 illustrates the two main types of beam expanders. The<br />
Keplerian type consists of two positive lenses which are positioned<br />
with their focal points nominally coincident. The Galilean type consists<br />
of a negative diverging lens, followed by a positive collimating<br />
lens, again positioned with their focal points nominally coincident.<br />
In both cases, the overall length of the optical system is given by<br />
overall length = f + f 1 2<br />
and the magnification is given by<br />
magnification = f f2<br />
1<br />
where a negative sign, in the Galilean system, indicates an inverted<br />
image (which is unimportant for laser beams). The Keplerian system,<br />
0.8 mm<br />
01 LDK 001<br />
01 LAO 059<br />
8 mm<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 2.12<br />
45 mm 80 m<br />
Lens spacing adjusted empirically to achieve the desired spot size at 80 m<br />
2.10 1 Visit Us OnLine! www.mellesgriot.com
Figure 2.13<br />
Keplerian beam expander<br />
f 1 f 2<br />
Galilean beam expander<br />
f 1<br />
f 2<br />
Two main types of beam expanders<br />
with its internal point of focus, allows one to utilize a spatial filter,<br />
while the Galilean system has the advantage of shorter length for<br />
a given magnification.<br />
In order to determine necessary focal lengths for an expander,<br />
we need to solve these two equations for the two unknowns.<br />
In this case,<br />
and<br />
f + f = 50<br />
1 2<br />
f<br />
f<br />
2<br />
1<br />
= 410.<br />
Using a negative value for the magnification will provide us<br />
with a Galilean expander. This yields values of f 2 = 55.5 mm and<br />
f 1 = 45.5 mm.<br />
Figure 2.14<br />
01 LDK 001<br />
01 LAO 059 01 LLP 017<br />
Ideally, a plano-concave diverging lens is used for minimum<br />
spherical aberration, but the shortest catalog focal length available is<br />
410 mm. There is, however, a biconcave lens with a focal length of<br />
45 mm (01 LDK 001). Even though this is not the optimum shape<br />
lens for this application, the extremely short focal length is likely to have<br />
negligible aberrations at this f-number. Ray tracing would confirm<br />
this.<br />
Now that we have selected a diverging lens with a focal length<br />
of 45 mm, we need to choose a collimating lens with a focal length<br />
of 50 mm. To determine whether a plano-convex lens is acceptable,<br />
check the spherical aberration formula:<br />
spot size resulting from spherical aberration<br />
0.067 × 50<br />
= =14 mm.<br />
3<br />
6.25<br />
The spot diameter resulting from diffraction is<br />
2w =<br />
0<br />
43<br />
2 (0.6328 × 10 ) 50<br />
= 5 mm.<br />
p4.0<br />
43<br />
0.6328 × 10 × 100<br />
w = = 50 mm.<br />
p0.4<br />
45 mm 95 mm<br />
Laser focusing system with long working distance<br />
Clearly, a plano-convex lens will not be adequate. The next choice<br />
would be an achromat, such as the 01 LAO 059. The data in the spot<br />
size charts on page 1.26 indicates that this lens is probably diffraction<br />
limited at this f-number. Our final system would therefore consist of<br />
the 01 LDK 001 spaced about 45 mm from the 01 LAO 059, which<br />
would have its flint element facing toward the laser.<br />
Example 2: Obtain 10 mm spot at > 100 mm<br />
Focus the output of an 05 LHR 151 to a spot diameter of 10 mm,<br />
but with the constraint that the last surface of the focusing optics<br />
is no closer than 100 mm to the focal point (see figure 2.14).<br />
Using a 100-mm-focal-length lens, the Gaussian beam focusing<br />
equation yields a spot radius of<br />
Thus, even a diffraction-limited focusing lens, with a 100-mm<br />
focal length, will produce a 100-µm-diameter focal spot with an<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 2.11
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
0.8-mm-diameter input beam. In order to achieve the spot size<br />
wanted, the beam must first be expanded by a factor of 10 before<br />
it is focused. The 10# expander described in the previous example<br />
could perform the task, as could any of the standard 10# expanders<br />
offered by Melles Griot.<br />
For focusing, we now have an 8-mm-diameter beam going into<br />
the 100-mm-focal-length lens, so we are operating at f/12.5. At this<br />
f-number we can probably use a plano-convex lens, but it is a good<br />
idea to check the spherical aberration to make sure.<br />
0.067 × 100<br />
spot size (spherical aberration) = = 3 mm.<br />
3<br />
12.5<br />
The plano-convex lens, oriented with its convex surface toward<br />
the beam expander, will provide diffraction-limited performance in<br />
this case.<br />
Although the effects of manufacturing tolerances should always<br />
be taken into account when choosing a standard catalog lens, they<br />
are not significant for the input lens of this beam expander because<br />
the aperture is so small. With a diameter of 1 mm or less, virtually<br />
any of the lenses in this catalog introduce only a fraction of a wave<br />
of wavefront distortion as a result of manufacturing errors. However,<br />
with a larger beam, lens quality is a consideration. One of the<br />
precision-grade lenses, in this case the 01 LLP 017, should be used<br />
for this precision application.<br />
Example 3: Collimate a diode laser<br />
Collect and collimate the output of a diode laser to a 25-mmdiameter<br />
diffraction-limited beam. The output wavelength is 780 nm<br />
and has a full-angle divergence of 60°!20° (see figure 2.15).<br />
The first step is to determine the numerical aperture needed to<br />
collect all the light from a source with a 60-degree divergence angle.<br />
Since numerical aperture is defined to be the sine of the half angle<br />
of divergence,<br />
NA = sin 30º = 0.5.<br />
Stated in terms of f-number, 1/(2 NA), this is f/1. At this low<br />
f-number we can immediately rule out virtually any simple lens or<br />
achromat; even if a simple lens were available at this low<br />
f-number, it would not provide the performance level required. The<br />
best choice would be a highly corrected, multielement diode laser<br />
collimating lens, such as the 06 GLC 002, which has a numerical<br />
aperture of 0.5.<br />
The 06 GLC 002 yields a collimated elliptical beam with dimensions<br />
of 8 mm ! 2.7 mm. The smaller dimension of this beam must<br />
be expanded to match the larger dimension; otherwise, it will have<br />
a larger beam divergence because of diffraction. Since there is<br />
approximately a 3:1 ratio in the two dimensions, we will use a 3#<br />
anamorphic prism pair, 06 GPA 004, to accomplish the expansion.<br />
This will now yield a collimated beam 8 mm in diameter.<br />
The next step is to expand the beam by a factor of 3.125#in order<br />
to get to the desired 25-mm beam diameter. Since no constraint has<br />
been given on the length of our optical system, we’ll play it safe and<br />
operate our beam expander at a minimum of f/10. This virtually<br />
ensures diffraction-limited performance, even with singlets.<br />
At f/10 and an 8-mm-diameter input beam, we would need a<br />
focal length of 80 mm for the input lens of our collimator. Since we<br />
are looking for diffraction-limited performance, our best choice<br />
would be one of the precision diode laser singlets (06 LXP series).<br />
Once again, we choose a high-precision lens because our beam has<br />
a fairly large diameter and the effects of manufacturing tolerances<br />
must be considered.<br />
The closest focal length we have in this series of lenses is the<br />
06 LXP 009 with a focal length of 110 mm. Operating at f/13.75,<br />
we will have diffraction-limited performance, which can be verified<br />
by using the formula for spherical aberration. We now need a<br />
collimating lens with a focal length of 3.125 ! 110 mm = 344 mm.<br />
The best choice is probably the 01 LAO 277 because there is no<br />
precision singlet lens with the necessary focal length. The achromat<br />
is also manufactured to tighter tolerances.<br />
The final system would then consist of the 06 GLC 002 mated<br />
directly to the 06 GPA 004, followed by the 06 LXP 009 with its<br />
curved surface facing toward the diode laser. The spacing between<br />
the 06 LXP 009 and 06 GPA 004 is not critical. Finally, the<br />
01 LAO 277 would follow, spaced approximately 455 mm from the<br />
singlet, with its flint surface facing toward the diode laser.<br />
Since the standard coating supplied with the 01 LAO series<br />
achromats does not perform very well at 780 nm, this lens should<br />
be specified with a /076 coating, which is optimized for performance<br />
at 780 nm.<br />
06 GPA 004<br />
06 LXP 009 01 LAO 277<br />
06 GLC 002<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 2.15<br />
1.1 mm<br />
455 mm<br />
Melles Griot diode laser components, showing how they may be used in relation to each other<br />
2.12 1 Visit Us OnLine! www.mellesgriot.com
<strong>Optical</strong> Specifications 3<br />
Wavefront Distortion 3.2<br />
Centration 3.3<br />
Modulation Transfer Function 3.4<br />
Cosmetic Surface Quality – U.S. Military Specifications 3.6<br />
Surface Accuracy 3.8<br />
3.1 1<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong>
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Wavefront Distortion<br />
Sometimes the best specification for an optical component is<br />
its effect on the emergent wavefront. This is particularly true for<br />
optical flats, collimation lenses, mirrors, and retroreflectors where<br />
the presumed effect of the element is to transmit or reflect the<br />
wavefront without changing its shape. Wavefront distortion is often<br />
characterized by the peak-to-valley deformation of the emergent<br />
wavefront from its intended shape. Specifications are normally<br />
quoted in fractions of a wavelength.<br />
Consider a perfectly plane, monochromatic wavefront, incident<br />
at an angle normal to the face of a window. Deviation from perfect<br />
surface flatness, as well as inhomogeneity of the bulk material<br />
refractive index of the window, will cause a deformation of the<br />
transmitted wavefront away from the ideal plane wave. In a<br />
retroreflector, each of the faces plus the material will affect the<br />
emergent wavefront. Consequently, any reflecting or refracting<br />
element can be characterized by the distortions imparted to a perfect<br />
incident wavefront.<br />
INTERFEROMETER MEASUREMENTS<br />
Melles Griot measures wavefront distortion with a laser<br />
interferometer. The wavefront from a helium neon laser<br />
(l = 632.8 nm) is expanded and then divided into a reference<br />
wavefront and test wavefronts by using a partially transmitting<br />
reference surface. The reference wavefront is reflected back to the<br />
interferometer, and the test wavefront is transmitted through the<br />
surfaces to the test element. The reference surface is a known flat<br />
or spherical surface whose surface error is on the order of l/20.<br />
When the test wavefront is reflected back to the interferometer,<br />
either from the surface being tested or from another l/20 reference<br />
surface, the reference and test wavefronts recombine at the<br />
interferometer. Constructive and destructive interference occurs<br />
between the two wavefronts. A difference in the optical paths of<br />
the two wavefronts is caused by any error present in the test element<br />
and any tilt of one wavefront relative to the other. The fringe pattern<br />
is projected onto a viewing screen or camera system.<br />
A slight tilt of the test wavefront to the reference wavefront produces<br />
a set of fringes whose parallelism and straightness depend on<br />
the element under test. The distance between successive fringes<br />
(usually measured from dark band to dark band) represents one<br />
wavelength difference in the optical path traveled by the two<br />
wavefronts. In surface and transmitted wavefront testing, the test<br />
wavefront travels through an error in the test piece twice. Therefore,<br />
one fringe spacing represents one half wavelength of surface<br />
error or transmission error of the test element.<br />
A determination of the convexity or concavity of the error in the<br />
test element can be made if the zero-order direction of the interference<br />
cavity (the space between the reference and test surfaces) is<br />
known. The zero-order direction is the direction of the center of tilt<br />
between the reference and test wavefronts.<br />
Fringes that curve around the center of tilt (zero-order) are<br />
convex, as a result of a high area on the test surface. Conversely,<br />
fringes that curve away from the center of tilt (zero-order) are<br />
concave as a result of a low area on the test surface.<br />
By using a known tilt and zero-order direction, the amount and<br />
direction (convex or concave) of the error in the test element can be<br />
determined from the fringe pattern. Six fringes of tilt are introduced<br />
for typical examinations. Melles Griot uses wavefront distortion<br />
measurements to characterize achromats, windows, filters, beamsplitters,<br />
prisms, and many other optical elements. This testing<br />
method is consistent with the way in which these components are<br />
normally used.<br />
INTERFEROGRAM INTERPRETATION<br />
Melles Griot tests lenses with a noncontact phase-measuring<br />
interferometer. The interferometer has a zoom feature to increase<br />
resolution of the optic under test. The interferometric cavity length<br />
is modulated, and a computerized data analysis program is used<br />
to interpret the interferogram. This computerized analysis increases<br />
the accuracy and repeatability of each measurement and eliminates<br />
subjective operator interpretation.<br />
3.2 1 Visit Us OnLine! www.mellesgriot.com
Centration<br />
The mechanical axis and optical axis exactly coincide in a<br />
perfectly centered lens.<br />
OPTICAL AND MECHANICAL AXES<br />
For a simple lens, the optical axis is defined as a straight line<br />
that joins the centers of lens curvature. For a plano-convex or planoconcave<br />
lens, the optical axis is the line through the center of<br />
curvature and perpendicular to the plano surface.<br />
The mechanical axis is determined by the way in which the lens<br />
will be mounted during use. There are typically two types of<br />
mounting configurations, edge mounting and surface mounting.<br />
With edge mounting, the mechanical axis is the centerline of the lens<br />
mechanical edge. Surface mounting uses one surface of the lens as<br />
the primary stability for lens tip and then encompasses the lens<br />
diameter for centering. The mechanical axis for this type of mounting<br />
is a line perpendicular to the mounting surface and centered on<br />
the entrapment diameter.<br />
Ideally, the optical and mechanical axes coincide. The tolerance<br />
on centration is the allowable amount of radial separation of these<br />
two axes, measured at the focal point of the lens. The centration<br />
angle is equal to the inverse tangent of the allowable radial separation<br />
divided by the focal length.<br />
MEASURING CENTRATION ERROR<br />
Centration error is measured by rotating the lens on its mechanical<br />
axis and observing the orbit of the focal point. To determine<br />
the centration error, the radius of this orbit is divided by the lens focal<br />
length and then converted to an angle.<br />
Figure 3.1<br />
optical axis<br />
mechanical axis<br />
C 2<br />
v<br />
H H″<br />
Centration and orbit of apparent focus<br />
c<br />
edge grinding removes<br />
material outside imaginary cylinder<br />
DOUBLETS AND TRIPLETS<br />
It is more difficult to achieve a given centration specification<br />
for a doublet than it is for a singlet because each element must be<br />
individually centered to a tighter specification, and the two optical<br />
axes must be carefully aligned during the cementing process.<br />
Centration is even more complex for triplets because three optical<br />
axes must be aligned. The centration error of doublets and triplets<br />
is measured in the same manner as that of simple lenses. One<br />
method used to obtain precise centration in compound lenses is<br />
to align the elements optically and edge the combination.<br />
CYLINDRICAL OPTICS<br />
Cylindrical optics can be evaluated for centering error in a<br />
manner similar to simple lenses. The major difference is that<br />
cylindrical optics have mechanical and optical planes rather than<br />
axes. The mechanical plane is established by the expected mounting,<br />
which can be edge only or the surface-edge combination<br />
described above. The radial separation between the focal line and<br />
the established mechanical plane is the centering error and can<br />
be converted into an angular deviation in the same manner as for<br />
simple lenses. The centering error is measured by first noting the focal<br />
line displacement in one orientation, then rotating the lens<br />
180 degrees and noting the new displacement. The centering error<br />
angle is the inverse tangent of the total separation divided by twice<br />
the focal length.<br />
H″′<br />
orbit of<br />
apparent focus<br />
true focus C 1<br />
F″<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 3.3
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Modulation Transfer Function<br />
The modulation transfer function (MTF), a quantitative measure<br />
of image quality, is far superior to any classic resolution criteria.<br />
MTF describes the ability of a lens or system to transfer object<br />
contrast to the image. Curves can be associated with the subsystems<br />
that make up a complete electro-optical or photographic system.<br />
MTF data can be used to determine the feasibility of overall system<br />
expectations.<br />
Bar-chart resolution testing of lens systems is deceptive because<br />
almost 20% of the energy arriving at a lens system from a bar chart<br />
is modulated at the third harmonic and higher frequencies. Consider<br />
instead a sine-wave chart in the form of a positive transparency in<br />
which transmittance varies in one dimension. Assume that the<br />
transparency is viewed against a uniformly illuminated background.<br />
The maximum and minimum transmittances are T max and T min ,<br />
respectively. A lens system under test forms a real image of the<br />
sine-wave chart, and the spatial frequency (u) of the image is<br />
measured in cycles per millimeter. Corresponding to the transmittances<br />
T max and T min are the image irradiances I max and I min .<br />
By analogy with Michelson’s definition of visibility of interference<br />
fringes, the contrast or modulation of the chart and image are<br />
defined, respectively, as<br />
and<br />
M =<br />
c<br />
M =<br />
i<br />
T<br />
T<br />
max<br />
max<br />
4 T<br />
+ T<br />
Imax<br />
4 I<br />
I + I<br />
max<br />
min<br />
min<br />
min<br />
min<br />
where M c is the modulation of the chart and M i is the modulation<br />
of the image.<br />
The modulation transfer function of the optical system at spatial<br />
frequency u is then defined to be<br />
MTF = MTF(u) = M /M i c .<br />
(3.1)<br />
(3.2)<br />
(3.3)<br />
The graph of MTF versus u is a modulation transfer function<br />
curve and is defined only for lenses or systems with positive focal<br />
length that form real images.<br />
It is often convenient to plot the magnitude of MTF (u) versus<br />
u. Changes in MTF curves are easily seen by graphical comparison.<br />
For example, for lenses, the MTF curves change with field<br />
angle positions and conjugate ratios. In a system with astigmatism<br />
or coma, different MTF curves are obtained that correspond to<br />
various azimuths in the image plane through a single image point.<br />
For cylindrical lenses, only one azimuth is meaningful. MTF<br />
curves can be either polychromatic or monochromatic. Polychromatic<br />
curves show the effect of any chromatic aberration that<br />
may be present. For a well-corrected achromatic system,<br />
polychromatic MTF can be computed by weighted averaging of<br />
monochromatic MTFs at a single image surface. MTF can also<br />
be measured by a variety of commercially available instruments.<br />
Most instruments measure polychromatic MTF directly.<br />
PERFECT CIRCULAR LENS<br />
The monochromatic, diffraction-limited MTF (or MDMTF) of<br />
a circular aperture (perfect aberration-free spherical lens) at an<br />
arbitrary conjugate ratio is given by the formula<br />
MDMTF(x) = 2 ⎡<br />
⎢arc cos (x) 4 x 14<br />
x<br />
π ⎣<br />
where the arc cosine function is in radians and x is the normalized<br />
spatial frequency defined by<br />
x = u<br />
u ic<br />
where u is the absolute spatial frequency and u ic is the incoherent<br />
diffraction cutoff spatial frequency. There are several formulas for<br />
u ic including<br />
u =<br />
ic<br />
=<br />
1.22<br />
r<br />
d<br />
⎛ 1.22l<br />
⎞<br />
n ′′ D 14⎜<br />
⎝ n ′′ D<br />
⎟<br />
⎠<br />
ls′′<br />
2<br />
⎛ 1.22l<br />
⎞<br />
2n ′′ sin(u ′′) 14⎜<br />
n D<br />
⎟<br />
⎝ ′′ ⎠<br />
=<br />
l<br />
2n ′′ sin(u ′′)<br />
=<br />
l<br />
= n ′′ D<br />
(3.6)<br />
ls′′<br />
where r d is the linear spot radius in the case of pure diffraction<br />
(Airy disc radius), D is the diameter of the lens clear aperture (or<br />
of a stop in near-contact), l is the wavelength, s″ is the secondary<br />
conjugate distance, u″ is the largest angle between any ray and the<br />
optical axis at the secondary conjugate point, the product n″ sin(u″)<br />
is by definition the image space numerical aperture, and n″ is the<br />
image space refractive index. It is essential that D, l, and s″ have<br />
consistent units (usually millimeters, in which case u and u ic will be<br />
in cycles per millimeter). The relationship<br />
sin(u ′′) = D<br />
2s′′<br />
implies that the secondary principal surface is a sphere centered<br />
upon the secondary conjugate point. This means that the lens is<br />
completely free of spherical aberration and coma, and, in the special<br />
case of infinite conjugate ratio (s″ = f″),<br />
u = n<br />
D ic ′′ . l f<br />
2<br />
2<br />
⎤<br />
⎥<br />
⎦<br />
(3.4)<br />
(3.5)<br />
(3.7)<br />
(3.8)<br />
3.4 1 Visit Us OnLine! www.mellesgriot.com
PERFECT RECTANGULAR LENS<br />
The MDMTF of a rectangular aperture (perfect aberrationfree<br />
cylindrical lens) at arbitrary conjugate ratio is given by the<br />
formula<br />
MDMTF(x) = (14<br />
x)<br />
where x is again the normalized spatial frequency u/u ic , where, in<br />
the present cylindrical case,<br />
u =<br />
ic<br />
1<br />
r<br />
d<br />
and r d is one-half the full width of the central stripe of the diffraction<br />
pattern measured from first maximum to first minimum. This<br />
formula differs by a factor of 1.22 from the corresponding formula<br />
in the circular aperture case. The following applies to both circular<br />
and rectangular apertures:<br />
u =<br />
ic<br />
2n ′′ sin(u ′′)<br />
.<br />
l<br />
The remaining three expressions for u ic in the circular aperture case<br />
can be applied to the present rectangular aperture case provided that<br />
two substitutions are made. Everywhere the constant 1.22 formerly<br />
appeared, it must be replaced by 1.00. Also, the aperture diameter<br />
D must now be replaced by the aperture width w. The relationship<br />
sin(u″) = w/2s″ means that the secondary principal surface is a<br />
circular cylinder centered upon the secondary conjugate line. In<br />
the special case of infinite conjugate ratio, the incoherent cutoff<br />
frequency for cylindrical lenses is<br />
u = n<br />
w ic ′′ . l f<br />
(3.9)<br />
(3.10)<br />
(3.11)<br />
(3.12)<br />
IDEAL PERFORMANCE AND REAL LENSES<br />
In an ideal lens, the x-intercept and the MDMTF-intercept are<br />
at unity (1.0). MDMTF(x) for the rectangular case is a straight line<br />
between these intercepts. For the circular case, MDMTF(x) is a<br />
curve that dips slightly below the straight line. These curves are<br />
shown in figure 3.2. Maximum contrast (unity) is apparent when<br />
spatial frequencies are low (i.e., for large features). Poor contrast is<br />
apparent when spatial frequencies are high (i.e., small features).<br />
All examples are limited at high frequencies by diffraction effects.<br />
A normalized spatial frequency of unity corresponds to the<br />
diffraction limit.<br />
All real cylindrical, monochromatic MTF curves fall on or below<br />
the straight MDMTF(x) line. Similarly, all real spherical and monochromatic<br />
MTF curves fall on or below the circular MDMTF(x)<br />
curve. Thus the two ideal MDMTF(x) curves represent the perfect<br />
(ideal) optical performance. <strong>Optical</strong> element or system quality is<br />
measured by how closely the real MTF curve approaches the<br />
corresponding ideal MDMTF(x) curve (see figure 3.3).<br />
MTF is an extremely sensitive measure of image degradation.<br />
To illustrate this, consider a lens having a quarter wavelength of<br />
spherical aberration. This aberration, barely discernible by eye,<br />
would reduce the MTF by as much as 0.2 at the midpoint of the<br />
spatial frequency range.<br />
MDMTF<br />
1.0<br />
.8<br />
.6<br />
.4<br />
.2<br />
circular aperture<br />
rectangular aperture<br />
0 .2 .4 .6 .8 1.0<br />
NORMALIZED SPATIAL FREQUENCY, X<br />
Figure 3.2 MDMTF(x) vs x, as a function of normalized<br />
spatial frequency, x<br />
MTF<br />
1.0<br />
.8<br />
.6<br />
.4<br />
.2<br />
lens with<br />
1 / 4 wavelength<br />
aberration<br />
MDMTF<br />
0 .2 .4 .6 .8 1.0<br />
NORMALIZED SPATIAL FREQUENCY, X<br />
Figure 3.3 MTF as a function of normalized spatial<br />
frequency, x<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 3.5
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Cosmetic Surface Quality —<br />
U.S. Military Specifications<br />
Cosmetic surface quality describes the level of defects that can<br />
be visually noted on the surface of an optical component. Specifically,<br />
it defines state of polish, freedom from scratches and digs, and<br />
edge treatment of components. These factors are important, not only<br />
because they affect the appearance of the component, but also<br />
because they scatter light, which adversely affects performance.<br />
Scattering can be particularly important in laser applications because<br />
of the intensity of the incident illumination. Unwanted diffraction<br />
patterns caused by scratches can lead to degraded system<br />
performance, and scattering of high-energy laser radiation can<br />
cause component damage. Overspecifying cosmetic surface quality,<br />
on the other hand, can be costly. Melles Griot components are<br />
tested at appropriate levels of cosmetic surface quality according<br />
to their intended application.<br />
The most common and widely accepted convention for specifying<br />
surface quality is the U.S. Military Surface Quality Specification,<br />
MIL-0-13830A, Amendment 3. The surface quality of all<br />
Melles Griot optics is tested in accordance with this specification.<br />
In Europe, an alternative specification, the DIN (Deutsche Industrie<br />
Norm) specification, DIN 3140, Sheet 7, is used. Melles Griot<br />
can also work to ISO-10110 requirements.<br />
SPECIFICATION STANDARDS<br />
As stated above, all optics in this catalog are referenced to MIL-<br />
0-13830A standards. These standards include scratches, digs, grayness,<br />
edge chips, and cemented interfaces. It is important to note that<br />
inspection of polished optical surfaces for scratches is accomplished<br />
by visual comparison to scratch standards. Thus, it is not the actual<br />
width of the scratch that is ascertained, but the appearance of the<br />
scratch as compared to these standards. A part is rejected if any<br />
scratches exceed the maximum size allowed. Digs, on the other hand,<br />
specified by actual defect size, can be measured quantitatively.<br />
Because of the subjective nature of this examination, it is critical<br />
to use trained inspectors who operate under standardized conditions<br />
in order to achieve consistent results. Melles Griot optics are<br />
compared by experienced quality assurance personnel using scratch<br />
and dig standards according to U.S. military drawing C7641866<br />
Rev L. Additionally, our inspection areas are equipped with lighting<br />
that meets the specific requirements of MIL-0-13830A.<br />
The scratch-and-dig designation for a component or assembly<br />
is specified by two numbers. The first defines allowable maximum<br />
scratch visibility, and the second refers to allowable maximum dig<br />
diameter, separated by a hyphen; for example,<br />
80–50 represents a commonly acceptable cosmetic standard.<br />
60–40 represents an acceptable standard for most scientific<br />
research applications.<br />
10–5 represents a precise standard for very demanding laser<br />
applications.<br />
SCRATCHES<br />
A scratch is defined as any marking or tearing of a polished<br />
optical surface. In principle, scratch numbers refer to the width<br />
of the reference scratch in ten thousandths of a millimeter. For<br />
example, an 80 scratch is equivalent to an 8-µm standard scratch.<br />
However, this equivalence is determined strictly by visual<br />
comparison, and the appearance of a scratch can depend upon the<br />
component material and the presence of any coatings. Therefore,<br />
a scratch on the test optic that appears equivalent to the 80 standard<br />
scratch is not necessarily 8 mm wide.<br />
If maximum visibility scratches are present (e.g., several<br />
60 scratches on a 60–40 lens), their combined lengths cannot exceed<br />
half of the part diameter. Even with some maximum visibility<br />
scratches present, MIL-0-13830A still allows many combinations<br />
of smaller scratch sizes and lengths on the polished surface.<br />
DIGS<br />
A dig is a pit or small crater on the polished optical surface.<br />
Digs are defined by their diameters, which are the actual sizes of the<br />
digs in hundredths of a millimeter. The diameter of an irregularly<br />
shaped dig is 1/2# (length plus width):<br />
50 dig = 0.5 mm in diameter<br />
40 dig = 0.4 mm in diameter<br />
30 dig = 0.3 mm in diameter<br />
20 dig = 0.2 mm in diameter<br />
10 dig = 0.1 mm in diameter.<br />
3.6 1 Visit Us OnLine! www.mellesgriot.com
The permissible number of maximum-size digs shall be one per<br />
each 20 mm of diameter (or fraction thereof) on any single surface.<br />
The sum of the diameters of all digs, as estimated by the inspector,<br />
shall not exceed twice the diameter of the maximum size specified<br />
per any 20-mm diameter. Digs less than 25 micrometers are ignored.<br />
EDGE CHIPS<br />
Lens edge chips are allowed only outside the clear aperture of<br />
the lens. The clear aperture is 90% of the lens diameter unless<br />
otherwise specified. Chips smaller than 0.5 mm are ignored, and<br />
those larger than 0.5 mm are ground so that there is no shine to<br />
the chip. The sum of the widths of chips larger than 0.5 mm cannot<br />
exceed 30% of the lens perimeter.<br />
Prism edge chips outside the clear aperture are allowed. If the<br />
prism leg dimension is 25.4 mm or less, chips may extend inward<br />
1.0 mm from the edge. If the leg dimension is larger than 25.4 mm,<br />
chips may extend inward 2.0 mm from the edge. Chips smaller than<br />
0.5 mm are ignored, and those larger than 0.5 mm must be stoned<br />
or ground, leaving no shine to the chip. The sum of the widths of<br />
chips larger than 0.5 mm cannot exceed 30% of the length of the edge<br />
on which they occur.<br />
CEMENTED INTERFACES<br />
Because a cemented interface is considered a lens surface, specified<br />
surface quality standards apply. Edge separation at a cemented<br />
interface cannot extend into the element more than half the distance<br />
to the element clear aperture up to a maximum of 1.0 mm. The sum<br />
of edge separations deeper than 0.5 mm cannot exceed 10% of the<br />
element perimeter.<br />
BEVELS<br />
Although bevels are not specified in MIL-0-13830A, our<br />
standard shop practice specifies that element edges are beveled to<br />
a face width of 0.25 to 0.5 mm at an angle of 45°±15°. Edges meeting<br />
at angles of 135° or larger are not beveled.<br />
COATING DEFECTS<br />
Defects caused by an optical element coating, such as scratches,<br />
voids, pinholes, dust, or stains, are considered with the scratchand-dig<br />
specification for that element. Coating defects are allowed<br />
if their size is within the stated scratch-and-dig tolerance. Coating<br />
defects are counted separately form substrate defects.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 3.7
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Surface Accuracy<br />
When attempting to specify how closely an optical surface<br />
conforms to its intended shape, a measure of surface accuracy is<br />
needed. Surface accuracy can be determined by interferometric<br />
techniques. Traditional techniques involve comparing the actual<br />
surface to a the test plate gage. In this approach, surface accuracy<br />
is measured by counting the number of rings or fringes and examining<br />
the regularity of the fringe. The accuracy of the fit between<br />
the lens and the test gage (as shown in figure 3.4) is described by the<br />
number of fringes seen when the gage is in contact with the lens. Test<br />
plates are made flat or spherical to within small fractions of a fringe.<br />
The accuracy of a test plate is only as good as the means used to<br />
measure its radii. Extreme care must be used when placing a test plate<br />
in contact with the actual surface to prevent damage to the surface.<br />
Modern techniques for measuring surface accuracy utilize phasemeasuring<br />
interferometry with advanced computer data analysis<br />
software. Removing operator subjectivity has made this approach<br />
considerably more accurate and repeatable. A zoom function can<br />
increase the resolution across the entire surface or a specific region<br />
to enhance the accuracy of the measurement.<br />
SURFACE FLATNESS<br />
Surface flatness is simply surface accuracy with respect to a<br />
plane reference surface. It is used extensively in mirror and optical<br />
flat specifications.<br />
POWER AND IRREGULARITY<br />
During manufacture, a precision component is frequently compared<br />
with a test plate that has an accurate polished surface that is<br />
the inverse of the surface under test. When the two surfaces are<br />
brought together and viewed in nearly monochromatic light,<br />
Newton’s rings (interference fringes caused by the near-surface<br />
standard surface in contact<br />
contact) appear. The number of rings indicates the difference in<br />
radius between the surfaces. This is known as power or sometimes<br />
as figure. It is measured in rings that are equivalent to half<br />
wavelengths.<br />
Beyond their number, the rings may exhibit distortion that<br />
indicates nonuniform shape differences. The distortion may be local<br />
to one small area, or it may be in the form of noncircular fringes<br />
over the whole aperture. All such nonuniformities are known<br />
collectively as irregularity.<br />
test surface<br />
maximum deviation<br />
air gap between surfaces<br />
reference surface<br />
surface accuracy<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 3.4<br />
Surface accuracy<br />
3.8 1 Visit Us OnLine! www.mellesgriot.com
Material Properties 4<br />
Material Properties Overview 4.2<br />
Introduction 4.3<br />
<strong>Optical</strong> Properties 4.4<br />
Mechanical and Chemical Properties 4.6<br />
Melles Griot Lens Materials 4.7<br />
Five Schott Glass Types 4.8<br />
Synthetic Fused Silica 4.11<br />
<strong>Optical</strong> Crown Glass 4.14<br />
Low-Expansion Borosilicate Glass 4.15<br />
Sapphire 4.16<br />
ZERODUR ® 4.17<br />
Calcium Fluoride 4.18<br />
4.1 1<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong>
Fundamental Optics<br />
Material Properties Overview<br />
Material<br />
Usable Transmission Range<br />
Index of<br />
Refraction<br />
Features<br />
BK7<br />
BK7<br />
1.52 @<br />
0.55 mm<br />
Excellent all-around lens material provides broad transmission with<br />
excellent mechanical characteristics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
LaSFN9<br />
SF11<br />
F2<br />
BaK1<br />
<strong>Optical</strong>-Quality<br />
Synthetic<br />
Fused Silica<br />
(OQSFS)<br />
UV-Grade<br />
Synthetic<br />
Fused Silica<br />
(UVGSFS)<br />
<strong>Optical</strong> Crown<br />
Glass<br />
Low-expansion<br />
borosilicate glass<br />
LEBG<br />
Sapphire<br />
Zinc Selenide<br />
Calcium<br />
Fluoride<br />
LaSFN9<br />
SF11<br />
F2<br />
BaK1<br />
OQSFS<br />
UVGSFS<br />
OPTICAL CROWN<br />
LEBG<br />
SAPPHIRE<br />
ZINC SELENIDE<br />
CALCIUM FLUORIDE<br />
1.86 @<br />
0.55 mm<br />
1.79 @<br />
0.55 mm<br />
1.62 @<br />
0.55 mm<br />
1.57 @<br />
0.55 mm<br />
1.46 @<br />
0.55 mm<br />
1.46 @<br />
0.55 mm<br />
1.52 @<br />
0.55 mm<br />
1.48 @<br />
0.55 mm<br />
1.77 @<br />
0.55 mm<br />
2.40 @<br />
10.6 mm<br />
1.399 @<br />
5 mm<br />
High-refractive-index flint glass provides more power with less<br />
curvature needed<br />
High-refractive-index flint glass provides more power with less<br />
curvature needed<br />
Material represents a good compromise between higher index and<br />
acceptable mechanical characteristics<br />
Excellent all-around lens material, but has weaker chemical<br />
characteristics than BK7<br />
Material provides good UV transmission and superior mechanical<br />
characteristics<br />
Material provides excellent UV transmission and superior mechanical<br />
characteristics<br />
This lower tolerance glass can be used as a mirror substrate or in noncritical<br />
applications<br />
Excellent thermal stability, low cost, and homogeneity makes LEBG useful<br />
for high-temperature windows, mirror substrates, and condenser lenses<br />
Excellent mechanical and thermal characteristics make it a superior<br />
window material<br />
Zinc selenide is most popular for transmissive IR optics, transmits<br />
visible and IR, and has low absorption in the red end of the spectrum<br />
This popular UV excimer laser material is used for windows, lenses,<br />
and mirror substrates<br />
0.1 0.5 1.0 5.0 10.0<br />
WAVELENGTHS IN mm<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
4.2 1 Visit Us OnLine! www.mellesgriot.com
Introduction<br />
Glass manufacturers provide hundreds of different glass types<br />
with differing optical transmissibility and mechanical strengths.<br />
Melles Griot has simplified the task of selecting the right material<br />
for an optical component by offering each of our standard components<br />
in a single material, or in a small range of materials best<br />
suited to typical applications.<br />
There are, however, two instances in which one might need to<br />
know more about optical materials: one might need to determine<br />
the performance of a catalog component in a particular application,<br />
or one might need specific information to select a material for a<br />
custom component. The information given in this chapter is intended<br />
to help those in such situations.<br />
The most important material properties to consider in regard to<br />
an optical element are as follows:<br />
$ Transmission versus wavelength<br />
$ Index of refraction<br />
$ Thermal characteristics<br />
$ Mechanical characteristics<br />
$ Chemical characteristics<br />
$ Cost.<br />
Transmission versus Wavelength<br />
A material must be transmissive at the wavelength of interest if<br />
it is to be used for a transmissive component. A transmission curve<br />
allows the optical designer to estimate the attenuation of light, at<br />
various wavelengths, caused by internal material properties. For<br />
mirror substrates, the attenuation may be of no consequence.<br />
Index of Refraction<br />
The index of refraction, as well as the rate of change of index with<br />
wavelength (dispersion), might require consideration. High-index<br />
materials allow the designer to achieve a given power with less<br />
surface curvature, typically resulting in lower aberrations. On the<br />
other hand, most high-index flint glasses have higher dispersions,<br />
resulting in more chromatic aberration in polychromatic applications.<br />
They also typically have poorer chemical characteristics than lower<br />
index crown glasses.<br />
Thermal Characteristics<br />
The thermal expansion coefficient can be particularly important<br />
in applications in which the part is subjected to high temperatures,<br />
such as high-intensity projection systems. This is also of concern<br />
when components must undergo large temperature cycles, such as<br />
in optical systems used outdoors.<br />
Mechanical Characteristics<br />
The mechanical characteristics of a material are significant in<br />
many areas. They can affect how easy it is to fabricate the material<br />
into shape, which affects product cost. Scratch resistance is important<br />
if the component will require frequent cleaning. Shock and vibration<br />
resistance are important for military, aerospace, or certain<br />
industrial applications. Ability to withstand high pressure differentials<br />
is important for windows used in vacuum chambers.<br />
Chemical Characteristics<br />
The chemical characteristics of a material, such as acid or stain<br />
resistance, can also affect fabrication and durability. As with mechanical<br />
characteristics, chemical characteristics should be taken into<br />
account for optics used outdoors or in harsh conditions.<br />
Cost<br />
Cost is almost always a factor to consider when specifying<br />
materials. Furthermore, the cost of some materials, such as UVgrade<br />
synthetic fused silica, increases sharply with larger diameters<br />
because of the difficulty in obtaining large pieces of the material.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.3
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
<strong>Optical</strong> Properties<br />
The most important optical properties of a material are its<br />
internal and external transmittances, surface reflectances, and<br />
refractive indices. The formulas that connect these variables in the<br />
on-axis case are presented below.<br />
TRANSMISSION<br />
External transmittance is the single-pass irradiance transmittance<br />
of an optical element. Internal transmittance is the single-pass irradiance<br />
transmittance in the absence of any surface reflection losses<br />
(i.e., transmittance of the material). External transmittance is of<br />
paramount importance when selecting optics for an image-forming<br />
lens system because external transmittance neglects multiple<br />
reflections between lens surfaces. Transmittance measured with an<br />
integrating sphere will be slightly higher. Let T e denote the desired<br />
external irradiance transmittance (see equation 4.1), T i the<br />
corresponding internal transmittance, t 1 the single-pass transmittance<br />
of the first surface, and t 2 the single-pass transmittance of<br />
the second surface:<br />
4mt<br />
T e = t1t2T i = t1t2e<br />
c<br />
where e is the base of the natural system of logarithms, m is the<br />
absorption coefficient of the lens material, and t c is the lens center<br />
thickness. This allows for the possibility that the lens surfaces might<br />
have unequal transmittances (for example, one is coated and the<br />
other is not). Assuming that both surfaces are uncoated,<br />
t t = 14<br />
2r + r<br />
1 2<br />
where<br />
⎛<br />
r = n 4 1 ⎞<br />
⎜<br />
⎝ n + 1<br />
⎟<br />
⎠<br />
2<br />
2<br />
is the single-surface single-pass irradiance reflectance at normal<br />
incidence as given by the Fresnel formula. The refractive index n must<br />
be known or calculated from the material dispersion formula<br />
(equation 4.6). These results are monochromatic. Both m and n are<br />
functions of wavelength.<br />
To calculate either T i or the T e for a lens at any wavelength of<br />
interest, first find the value of absorption coefficient m (equation 4.4).<br />
Typically, internal transmittance T i is tabulated as a function of wavelength<br />
for two distinct thicknesses T c1 and T c2 , and m must be found<br />
from these. Thus<br />
1 ⎡1n T i (t c1)<br />
1n T i (t c2)<br />
⎤<br />
m = 4 ⎢ + ⎥<br />
2 ⎣⎢<br />
t c1<br />
t c2 ⎦⎥<br />
where the bar denotes averaging. In portions of the spectrum where<br />
absorption is strong, a value for T i is typically given only for the lesser<br />
thickness. Then<br />
1<br />
m = 4 1n T i .<br />
t<br />
c<br />
(4.1)<br />
(4.2)<br />
(4.3)<br />
(4.4)<br />
(4.5)<br />
When it is necessary to find transmittance at wavelengths other<br />
than those for which T i is tabulated, use linear interpolation.<br />
The on-axis T e value is normally the most useful, but some<br />
applications require that transmittance be known along other ray<br />
paths, or that it be averaged over the entire lens surface. The method<br />
outlined above is easily extended to encompass such cases. Values<br />
of t 1 and t 2 must be found from complete Fresnel formulas for arbitrary<br />
angles of incidence. The angles of incidence will be different<br />
at the two surfaces; therefore, t 1 and t 2 will generally be unequal.<br />
Distance t c , which becomes the surface-to-surface distance along<br />
a particular ray, must be determined by ray tracing. It is necessary<br />
to account separately for the s- and p-planes of polarization, and<br />
it is usually sufficient to average results for both planes at the end<br />
of the calculation.<br />
REFRACTIVE INDEX AND DISPERSION<br />
The Schott <strong>Optical</strong> Glass catalog offers nearly 300 different<br />
optical glasses. For lens designers, the most important difference<br />
among these glasses is the index of refraction and dispersion (rate<br />
of change of index with wavelength). Typically, an optical glass is<br />
specified by its index of refraction at a wavelength in the middle of<br />
the visible spectrum, usually 587.56 nm (the helium d-line), and by<br />
the Abbé v-value, defined to be v d = (n d 41)/ (n F 4n C ). The designations<br />
F and C stand for 486.1 nm and 656.3 nm, respectively. Here,<br />
v d shows how the index of refraction varies with wavelength. The<br />
smaller v d is, the faster the rate of change is. Glasses are roughly<br />
divided into two categories: crowns and flints. Crown glasses are<br />
those with n d < 1.60 and v d > 55, or n d > 1.60 and v d > 50. The<br />
others are flint glasses.<br />
The refractive index of glass from 365 to 2300 nm can be<br />
calculated by using the following formula:<br />
n =<br />
⎛ 2<br />
2<br />
2<br />
B1l<br />
C + B2l<br />
C + B3l<br />
C + 1<br />
⎞<br />
⎜ 2<br />
2<br />
2 ⎟<br />
⎝ l 4 1 l 4 2 l 4 3 ⎠<br />
Here l, the wavelength, must be in micrometers, and the constants<br />
B 1 through C 3 are given by the glass manufacturer. Our tabulation<br />
of these constants for the glasses used in our catalog<br />
components are presented on page 4.8. Values for other glasses<br />
can be obtained from the manufacturer’s literature. This equation<br />
yields an index value that is accurate to better than 1!10 45 over<br />
the entire transmission range, and even less in the visible spectrum.<br />
OTHER OPTICAL CHARACTERISTICS<br />
Homogeneity within Melt<br />
Homogeneity within melt is the amount of refractive index<br />
variation within the manufactured glass blank. Inhomogeneity of<br />
refractive index can result in transmitted wavefront distortion. The<br />
1/2<br />
(4.6)<br />
4.4 1 Visit Us OnLine! www.mellesgriot.com
maximum value for homogeneity within melt for all Schott optical<br />
glasses used in Melles Griot catalog components is 1!10 44 .<br />
Striae Grade<br />
Striae are thread-like inclusions within an optical glass. Striae<br />
grades are specified in U.S. military specification MIL-G 174B. All<br />
Melles Griot catalog components that utilize Schott optical glass<br />
are specified to have striae that conform to MIL-G 174B grade A.<br />
Grade A means that no visible striae, streaks, or cords are present<br />
in the glass.<br />
Stress Birefringence<br />
Mechanical stress in optical glass leads to birefringence (anisotropy<br />
in index of refraction) which can impair the optical performance of<br />
a finished component. <strong>Optical</strong> glass is annealed (heated and cooled)<br />
to remove any residual stress left over from the original manufacturing<br />
process. Schott Glass defines fine annealed glass to have a<br />
maximum of 12 nm/cm of residual stress birefringence for blanks<br />
of up to 800 mm in diameter and 100 mm in thickness.<br />
APPLICATION NOTE<br />
Fused-Silica Optics<br />
Synthetic fused silica, described on page 4.11, is an<br />
ideal optical material for many laser applications.<br />
It is transparent from as low as 180 nm to over<br />
2.0 mm, has low coefficient of thermal expansion,<br />
and is resistant to scratching and thermal shock.<br />
For more information on some of the specific<br />
components manufactured from fused silica, see the<br />
following pages: Lenses, pages 6.22–6.29; Mirrors,<br />
9.12–9.17; Beamsplitters, 11.4–11.8.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.5
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Mechanical and Chemical Properties<br />
Mechanical and chemical properties of glass are important to lens<br />
manufacturers. These properties can also be significant to the user,<br />
especially when the component will be used in a harsh environment.<br />
Different polishing techniques and special handling may be needed<br />
depending on whether the glass is hard or soft, or whether it is<br />
extremely sensitive to acid or alkali.<br />
To quantify the chemical properties of glasses, each glass is rated<br />
according to four categories: climatic resistance, stain resistance, acid<br />
resistance, and alkali and phosphate resistance.<br />
Climatic Resistance<br />
Humidity can cause a cloudy film to appear on the surface of<br />
some optical glass. Climatic resistance expresses the susceptibility<br />
of a glass to this process. In this test, glass is placed in a watervapor-saturated<br />
environment and subjected to a temperature cycle<br />
which alternately causes condensation and evaporation. The glass<br />
is given a rating from 1 to 4 depending on the amount of surface<br />
scattering induced by the test. A rating of 1 indicates little or no<br />
change after seven days of exposure; a rating of 4 means a significant<br />
change occurred in less than 30 hours.<br />
Stain Resistance<br />
Stain resistance expresses resistance to mildly acidic water<br />
solutions, such as fingerprints or perspiration. In this test, a few<br />
drops of a mild acid are placed on the glass. A colored stain, caused<br />
by interference, will appear if the glass starts to decompose. A rating<br />
from 1 to 5 is given to each glass, depending on how much time<br />
elapses before stains occur. A rating of 1 indicates no observed stain<br />
in 100 hours of exposure; a rating of 5 means that staining occurred<br />
in less than 0.2 hours.<br />
Acid Resistance<br />
Acid resistance quantifies the resistance of a glass to stronger<br />
acidic solutions. Acid resistance can be particularly important to<br />
lens manufacturers because acidic solutions are typically used to strip<br />
coatings from glass or to separate cemented elements. A rating<br />
from 1 to 4 indicates progressively less resistance to a pH 0.3 acid<br />
solution, and values from 51 to 53 are used for glass with too little<br />
resistance to be tested with such a strong solution.<br />
Alkali and Phosphate Resistance<br />
Alkali resistance is also important to the lens manufacturer<br />
since the polishing process usually takes place in an alkaline<br />
solution. Phosphate resistance is becoming more significant as<br />
users move away from cleaning methods that involve chlorofluorocarbons<br />
(CFCs) to those that may be based on traditional<br />
phosphate-containing detergents. In each case, a two-digit number<br />
is used to designate alkali or phosphate resistance. The first number,<br />
from 1 to 4, indicates the length of time that elapses before any<br />
surface change occurs in the glass, and the second digit reveals the<br />
extent of the change.<br />
Microhardness<br />
The most important mechanical property of glass is microhardness.<br />
A precisely specified diamond scribe is placed on the glass<br />
surface under a known force. The indentation is then measured.<br />
The Knoop and the Vickers microhardness tests are used to measure<br />
the hardness of a polished surface and a freshly fractured surface,<br />
respectively.<br />
APPLICATION NOTE<br />
Glass Manufacturers<br />
The catalogs of optical glass manufacturers contain<br />
products covering a very wide range of optical<br />
characteristics. However, it should be kept in mind<br />
that the glass types that exhibit the most desirable<br />
properties in terms of index of refraction and<br />
dispersion often have the least practical chemical and<br />
mechanical characteristics. Furthermore, poor<br />
chemical and mechanical attributes translate directly<br />
into increased component costs because working<br />
these sensitive materials increases fabrication time<br />
and lowers yield. Please contact us before specifying<br />
an exotic glass in an optical design so that we can<br />
advise you of the impact that that choice will have on<br />
part fabrication.<br />
4.6 1 Visit Us OnLine! www.mellesgriot.com
Melles Griot Lens Materials<br />
Melles Griot simple lenses are made of synthetic fused silica,<br />
BK7 grade A fine annealed glass, and several other materials. The<br />
following table identifies the materials used in Melles Griot lenses.<br />
Some of these materials are also used in prisms, mirror substrates,<br />
and other products.<br />
Glass type designations and physical constants are the same as<br />
those published by Schott Glass. Melles Griot occasionally uses<br />
corresponding glasses made by other glass manufacturers but only<br />
when this does not result in a significant change in optical properties.<br />
The quality of performance of optical lenses and prisms depends<br />
on the quality of the material used. No amount of skill during<br />
manufacture can eradicate striae, bubbles, inclusions, or variations<br />
in index. Melles Griot takes considerable care in its material selection,<br />
using only first-class optical materials from reputable glass manufacturers.<br />
The result is reliable, repeatable, consistent performance.<br />
The following physical constant values are reasonable averages<br />
based on historical experience. Individual material specimens may<br />
deviate from these means. Materials having tolerances more<br />
restrictive than those published in the rest of this chapter, or materials<br />
traceable to specific manufacturers, are available only on special<br />
request.<br />
BK7 OPTICAL GLASS<br />
A borosilicate crown glass, BK7, is the material used in many<br />
Melles Griot products. BK7 performs well in chemical tests so that<br />
special treatment during polishing is not necessary. BK7, relatively<br />
hard glass, does not scratch easily and can be handled without special<br />
precautions. The bubble and inclusion content of BK7 is very<br />
low: the bubble and inclusion content cross-section totals less than<br />
0.029 mm 2 per 100 cm 3 . Another important characteristic of BK7<br />
is its excellent transmittance, as low as 350 nm. Because of these properties,<br />
BK7 is used widely throughout the optics industry. A variant<br />
of BK7, designated UBK7, has transmission almost as low as<br />
300 nm. This special glass is useful in applications requiring a high<br />
index of refraction, the desirable chemical properties of BK7, and<br />
transmission deeper into the ultraviolet range.<br />
Melles Griot Lens Materials<br />
Materials<br />
Synthetic Fused Silica, UV Grade<br />
Synthetic Fused Silica, <strong>Optical</strong> Quality<br />
BK7, Grade A Fine Annealed<br />
LaSF N9, Grade A Fine Annealed<br />
BaK1, Grade A Fine Annealed<br />
<strong>Optical</strong> Crown<br />
Low-Expansion Borosilicate Glass<br />
(LEBG)<br />
SF11, Grade A Fine Annealed<br />
SK11 and SF5, Grade A Fine Annealed<br />
Sapphire<br />
Zinc Selenide<br />
Various Glass Combinations<br />
Melles Griot reserves the right to make material changes or substitutions on any optical components without prior notice.<br />
Lens Product Numbers<br />
01 LQC 01 LQP<br />
01 LQD 01 LQS<br />
01 LQB 01 LQT<br />
01 LQF<br />
Selected 01 CMP series<br />
01 LCN 01 LMN<br />
01 LCP 01 LMP<br />
01 LDK 01 LPK<br />
01 LDX 01 LPX<br />
01 LFS<br />
01 LPX 401 01 LPX 411<br />
01 LPX 405 01 LPX 413<br />
01 LPX 407 06 LMS<br />
01 LPX 415 01 LPX 423<br />
01 LPX 421<br />
01 LAG<br />
Selected 01 CMP series<br />
06 LXP<br />
06 LAI<br />
01 LSX<br />
12 LNZ 12 LPZ<br />
01 LAL 04 EWR 001<br />
01 LAO 04 OAS<br />
01 LAT 04 OAP<br />
01 LBX 06 DDL<br />
04 ECW 06 DBF<br />
04 EHY 06 GLC<br />
04 EPP 06 GLR<br />
04 ERA 001 09 LBM<br />
04 EWA 09 LCM<br />
04 EWP 001 09 LSL<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.7
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Five Schott Glass Types<br />
The following tables list the most important optical and physical<br />
constants for Schott optical glass types BK7, SF11, LaSFN9,<br />
BaK1, and F2. These types are used in most Melles Griot simple<br />
lens products and prisms. Index of refraction and transmission, as well<br />
as the most commonly required chemical characteristics and mechanical<br />
constants, are listed. Further numerical data and a more detailed<br />
discussion of the various testing processes can be found in the Schott<br />
<strong>Optical</strong> Glass catalog.<br />
The index of refraction data were obtained by using the constants<br />
listed below together with the dispersion formula (equation 4.6).<br />
The constants were determined through the index-of-refraction<br />
measurements of a typical melt for each glass type. Note that the<br />
dispersion formula is valid only within the wavelength range<br />
Physical Constants of Five Schott Glasses<br />
Melt-to-Melt Mean Index Tolerance<br />
Homogeneity within Melt<br />
Striae Grade (MIL-G-174-A)<br />
Stress Birefringence, nm/cm, Yellow Light<br />
Abbé Factor (v d )<br />
Constants of Dispersion Formula:<br />
B 1<br />
B 2<br />
B 3<br />
C 1<br />
C 2<br />
C 3<br />
Density (g /cm 43 )<br />
Coefficient of Linear Thermal Expansion (a):<br />
430º to +70º (per ºC)<br />
+20º to +300º (per ºC)<br />
Transformation Temperature<br />
Young’s Modulus (dynes/mm 2 )<br />
Climate Resistance<br />
Stain Resistance<br />
Acid Resistance<br />
Alkali Resistance<br />
Phosphate Resistance<br />
Knoop Hardness<br />
Poisson’s Ratio<br />
Glass Type<br />
BK7 SF11 LaSFN9 BaK1 F2<br />
±0.001<br />
±1!10 44<br />
A<br />
10<br />
64.17<br />
1.03961212<br />
2.31792344!10 41<br />
1.01046945<br />
6.00069867!10 43<br />
2.00179144!10 42<br />
1.03560653!10 2<br />
2.51<br />
7.1!10 46<br />
8.3!10 46<br />
557ºC<br />
8.20!10 9<br />
2<br />
0<br />
1.0<br />
2.0<br />
2.3<br />
610<br />
0.206<br />
±0.001<br />
±1!10 44<br />
A<br />
10<br />
25.76<br />
1.73848403<br />
3.11168974!10 41<br />
1.17490871<br />
1.36068604!10 42<br />
6.15960463!10 42<br />
1.21922711!10 2<br />
4.74<br />
6.1!10 46<br />
6.8!10 46<br />
505ºC<br />
6.60!10 9<br />
1<br />
0<br />
1.0<br />
1.2<br />
1.0<br />
450<br />
0.235<br />
listed. It can be used to interpolate refractive index at other wavelengths<br />
within this range (to a precision of 1!10 45 or better), but it<br />
should not be used to extrapolate to wavelengths beyond this range.<br />
Furthermore, the actual melt-to-melt tolerance on the index of refraction<br />
typically is about ±0.001.<br />
The internal transmittance values shown are melt-to-melt experimental<br />
means and may be affected by thermal history (coating,<br />
annealing, or tempering operations) after manufacture.<br />
For more detailed information of these materials, please refer to<br />
the Schott <strong>Optical</strong> Glass catalog.<br />
±0.002<br />
±1!10 44<br />
A<br />
10<br />
32.17<br />
1.97888194<br />
3.20435298!10 41<br />
1.92900751<br />
1.18537266!10 42<br />
5.27381770!10 42<br />
1.66256540!10 2<br />
4.44<br />
7.4!10 46<br />
8.4!10 46<br />
703ºC<br />
1.09!10 10<br />
2<br />
0<br />
2.0<br />
1.0<br />
1.0<br />
630<br />
0.286<br />
±0.001<br />
±1!10 44<br />
A<br />
10<br />
57.55<br />
1.12365662<br />
3.09276848!10 41<br />
8.81511957!10 41<br />
6.44742752!10 43<br />
2.22284402!10 42<br />
1.07297751!10 2<br />
3.19<br />
7.6!10 46<br />
8.6!10 46<br />
592ºC<br />
7.30!10 9<br />
2<br />
1<br />
3.3<br />
1.2<br />
2.0<br />
530<br />
0.252<br />
±0.001<br />
±1!10 44<br />
A<br />
10<br />
36.37<br />
1.34533359<br />
2.09073176!10 41<br />
9.37357162!10 41<br />
9.97743871!10 43<br />
4.70450767!10 42<br />
1.11886764!10 2<br />
3.61<br />
8.2!10 46<br />
9.2!10 46<br />
438ºC<br />
5.70!10 9<br />
1<br />
0<br />
1.0<br />
2.3<br />
1.3<br />
420<br />
0.220<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
4.8 1 Visit Us OnLine! www.mellesgriot.com
Refractive Index of Five Schott Glass Types<br />
Wavelength<br />
l<br />
Refractive Index, n<br />
Fraunhofer<br />
(nm) BK7 SF11 LaSFN9 BaK1 F2<br />
Designation Source Spectral Region<br />
351.1<br />
1.53894 — — 1.60062 1.67359<br />
Ar laser<br />
UV<br />
363.8<br />
1.53649 — — 1.59744 1.66682<br />
Ar laser<br />
UV<br />
404.7<br />
1.53024 1.84208 1.89844 1.58941 1.65064<br />
h<br />
Hg arc<br />
Violet<br />
435.8<br />
1.52668 1.82518 1.88467 1.58488 1.64202<br />
g<br />
Hg arc<br />
Blue<br />
441.6<br />
1.52611 1.82259 1.88253 1.58415 1.64067<br />
HeCd laser<br />
Blue<br />
457.9<br />
1.52461 1.81596 1.87700 1.58226 1.63718<br />
Ar laser<br />
Blue<br />
465.8<br />
1.52395 1.81307 1.87458 1.58141 1.63564<br />
Ar laser<br />
Blue<br />
472.7<br />
1.52339 1.81070 1.87259 1.58071 1.63437<br />
Ar laser<br />
Blue<br />
476.5<br />
1.52309 1.80946 1.87153 1.58034 1.63370<br />
Ar laser<br />
Blue<br />
480.0<br />
1.52283 1.80834 1.87059 1.58000 1.63310<br />
F′<br />
Cd arc<br />
Blue<br />
486.1<br />
1.52238 1.80645 1.86899 1.57943 1.63208<br />
F<br />
H 2 arc<br />
Blue<br />
488.0<br />
1.52224 1.80590 1.86852 1.57927 1.63178<br />
Ar laser<br />
Blue<br />
496.5<br />
1.52165 1.80347 1.86645 1.57852 1.63046<br />
Ar laser<br />
Green<br />
501.7<br />
1.52130 1.80205 1.86524 1.57809 1.62969<br />
Ar laser<br />
Green<br />
514.5<br />
1.52049 1.79880 1.86245 1.57707 1.62790<br />
Ar laser<br />
Green<br />
532.0<br />
1.51947 1.79479 1.85901 1.57580 1.62569<br />
Nd laser<br />
Green<br />
546.1<br />
1.51872 1.79190 1.85651 1.57487 1.62408<br />
e<br />
Hg arc<br />
Green<br />
587.6<br />
1.51680 1.78472 1.85025 1.57250 1.62004<br />
d<br />
He arc<br />
Yellow<br />
589.3<br />
1.51673 1.78446 1.85002 1.57241 1.61989<br />
D<br />
Na arc<br />
Yellow<br />
632.8<br />
1.51509 1.77862 1.84489 1.57041 1.61656<br />
HeNe laser<br />
Red<br />
643.8<br />
1.51472 1.77734 1.84376 1.56997 1.61582<br />
C′<br />
Cd arc<br />
Red<br />
656.3<br />
1.51432 1.77599 1.84256 1.56949 1.61503<br />
C<br />
H 2 arc<br />
Red<br />
694.3<br />
1.51322 1.77231 1.83928 1.56816 1.61288<br />
Ruby laser<br />
Red<br />
786.0<br />
1.51106 1.76558 1.83323 1.56564 1.60889<br />
IR<br />
821.0<br />
1.51037 1.76359 1.83142 1.56485 1.60768<br />
IR<br />
830.0<br />
1.51020 1.76311 1.83098 1.56466 1.60739<br />
GaAlAs laser<br />
IR<br />
852.1<br />
1.50980 1.76200 1.82997 1.56421 1.60671<br />
s<br />
Ce arc<br />
IR<br />
904.0<br />
1.50893 1.75970 1.82785 1.56325 1.60528<br />
GaAs laser<br />
IR<br />
1014.0<br />
1.50731 1.75579 1.82420 1.56152 1.60279<br />
t<br />
Hg arc<br />
IR<br />
1060.0<br />
1.50669 1.75445 1.82293 1.56088 1.60190<br />
Nd laser<br />
IR<br />
1300.0<br />
1.50370 1.74901 1.81764 1.55796 1.59813<br />
InGaAsP laser<br />
IR<br />
1500.0<br />
1.50127 1.74554 1.81412 1.55575 1.59550<br />
IR<br />
1550.0<br />
1.50065 1.74474 1.81329 1.55520 1.59487<br />
IR<br />
1970.1<br />
1.49495 1.73843 1.80657 1.55032 1.58958<br />
Hg arc<br />
IR<br />
2325.4<br />
1.48921 1.73294 1.80055 1.54556 1.58465<br />
Hg arc<br />
IR<br />
Visit Us Online! www.mellesgriot.com 1 4.9<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong>
Fundamental Optics<br />
Internal Transmittance of Five Schott Glass Types<br />
Internal Transmittance (%)<br />
Wavelength<br />
l<br />
BK7<br />
Thickness (mm)<br />
SF11<br />
Thickness (mm)<br />
LaSFN9<br />
Thickness (mm)<br />
BaK1<br />
Thickness (mm)<br />
F2<br />
Thickness (mm)<br />
(nm) 5 25 5 25 5 25 5 25 5 25<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
300<br />
310<br />
320<br />
330<br />
340<br />
350<br />
360<br />
370<br />
380<br />
390<br />
400<br />
420<br />
440<br />
460<br />
480<br />
500<br />
540<br />
580<br />
620<br />
660<br />
700<br />
0.26<br />
0.59<br />
0.81<br />
0.91<br />
0.96<br />
0.986<br />
0.991<br />
0.995<br />
0.996<br />
0.998<br />
0.998<br />
0.998<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
—<br />
0.07<br />
0.35<br />
0.65<br />
0.83<br />
0.93<br />
0.96<br />
0.974<br />
0.980<br />
0.989<br />
0.991<br />
0.993<br />
0.994<br />
0.994<br />
0.995<br />
0.996<br />
0.996<br />
0.996<br />
0.997<br />
0.997<br />
0.998<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
0.13<br />
0.46<br />
0.73<br />
0.93<br />
0.97<br />
0.986<br />
0.991<br />
0.995<br />
0.998<br />
0.998<br />
0.998<br />
0.999<br />
0.999<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
0.02<br />
0.21<br />
0.69<br />
0.86<br />
0.93<br />
0.95<br />
0.976<br />
0.988<br />
0.992<br />
0.992<br />
0.993<br />
0.994<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
0.55<br />
0.70<br />
0.80<br />
0.86<br />
0.92<br />
0.94<br />
0.96<br />
0.972<br />
0.980<br />
0.990<br />
0.995<br />
0.996<br />
0.997<br />
0.997<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
—<br />
0.05<br />
0.18<br />
0.34<br />
0.47<br />
0.66<br />
0.76<br />
0.83<br />
0.87<br />
0.91<br />
0.95<br />
0.975<br />
0.983<br />
0.986<br />
0.990<br />
0.64<br />
0.81<br />
0.89<br />
0.94<br />
0.97<br />
0.981<br />
0.990<br />
0.995<br />
0.996<br />
0.997<br />
0.998<br />
0.998<br />
0.998<br />
0.998<br />
0.998<br />
0.998<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.11<br />
0.34<br />
0.56<br />
0.73<br />
0.84<br />
0.91<br />
0.95<br />
0.976<br />
0.982<br />
0.987<br />
0.988<br />
0.989<br />
0.989<br />
0.990<br />
0.991<br />
0.991<br />
0.993<br />
0.994<br />
0.995<br />
0.996<br />
0.997<br />
—<br />
—<br />
—<br />
—<br />
0.81<br />
0.95<br />
0.973<br />
0.987<br />
0.992<br />
0.995<br />
0.996<br />
0.997<br />
0.998<br />
0.998<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
0.999<br />
—<br />
—<br />
—<br />
—<br />
0.42<br />
0.78<br />
0.87<br />
0.94<br />
0.96<br />
0.973<br />
0.982<br />
0.987<br />
0.989<br />
0.991<br />
0.992<br />
0.993<br />
0.995<br />
0.995<br />
0.995<br />
0.995<br />
0.996<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
4.10 1 Visit Us OnLine! www.mellesgriot.com
Synthetic Fused Silica<br />
Fused silica is an ideal optical material for many applications. It<br />
is transparent over a wide spectral range, has a low coefficient of<br />
thermal expansion, and is resistant to scratching and thermal shock.<br />
Synthetic fused silica (amorphous silicon dioxide) is formed by<br />
chemical combination of silicon and oxygen. It is not to be confused<br />
with fused quartz, which is made by crushing and melting natural<br />
crystals, or by fusing silica sand, which results in a granular microstructure<br />
and bubble entrapment. Microstructure and impurities<br />
lead to local index variations and contribute, along with bubbles and<br />
opaque particles, to reduced transmission throughout the spectrum.<br />
Synthetic fused silica is far purer than fused quartz. This<br />
increased purity ensures higher ultraviolet transmission and freedom<br />
from striae or inclusions. The synthetic fused-silica materials used<br />
by Melles Griot are manufactured by flame hydrolysis to extremely<br />
high standards. The resultant material is colorless and non-crystalline,<br />
and it has an impurity content of only about one part per million.<br />
Controlling the purity of reactants and the conditions of reaction<br />
ensures the high quality of the synthetic fused silica from which our<br />
lenses are made.<br />
Synthetic fused-silica lenses offer a number of advantages over<br />
glass or fused quartz:<br />
$ Greater ultraviolet and infrared transmission<br />
$ Low coefficient of thermal expansion, which provides stability<br />
and resistance to thermal shock over large temperature<br />
excursions<br />
$ Wider thermal operating range<br />
$ Increased hardness and resistance to scratching<br />
$ Much higher resistance to radiation darkening from<br />
ultraviolet, X-rays, gamma rays, and neutrons.<br />
<strong>Optical</strong>-quality synthetic fused silica (OQSFS) lenses are ideally<br />
suited for applications in energy-gathering and imaging systems<br />
in the mid-ultraviolet, visible, and near-infrared spectral regions.<br />
The low dispersion of fused silica reduces chromatic aberration.<br />
UV-grade synthetic fused silica (UVGSFS) is selected to offer<br />
the highest transmission (especially in the deep ultraviolet) and very<br />
low fluorescence levels (approximately 0.1% that of fused natural<br />
quartz excited at 254 nm). UV-grade synthetic fused silica does not<br />
fluoresce in response to wavelengths longer than 290 nm. In deep<br />
ultraviolet applications, UV-grade synthetic fused silica is an ideal<br />
choice. Its tight index tolerance ensures highly predictable lens<br />
specifications.<br />
The left-hand table on page 4.13 shows the refractive index of a<br />
typical UV-grade synthetic fused silica versus wavelength at 20ºC.<br />
To obtain the index for optical-quality synthetic fused silica, round<br />
the values off to the fourth decimal place.<br />
Glass transmittances are affected by thermal history after manufacture,<br />
as well as during the manufacturing process. Depending on<br />
the manufacturer and subsequent thermal processing (coating,<br />
annealing, or tempering), it is possible for any optical glass, including<br />
BK7, to show internal transmittance reductions of several percent<br />
across the entire spectrum with external transmittance correspondingly<br />
affected. Transmittance of all glass is especially uncertain at<br />
wavelengths approaching the water absorption band at 2.7 mm.<br />
Synthetic fused silica also shows batch-to-batch transmittance<br />
variations, especially in deep ultraviolet and infrared. These<br />
variations are related to manufacture and impurity content rather<br />
than subsequent history. In the ultraviolet, these variations have<br />
been attributed to uncontrollable fluctuations in metallic impurity<br />
content at the parts per billion level. Ultraviolet transmittance is the<br />
basis for the classifications UV grade and optical quality. A<br />
specification of UV grade ensures that a specimen is represented by<br />
the broadest curve. Transmittance curves for optical quality may fall<br />
anywhere between the UVGSFS curve and the OQSFS curve shown<br />
in figure 4.1.<br />
Infrared batch-to-batch transmittance variations in synthetic<br />
fused silica are attributable to fluctuations in the OH chemical bond<br />
content. These variations are most pronounced at wavelengths near<br />
and beyond the water absorption band at 2.7 mm and are normally<br />
uncontrolled because ultraviolet transmittance is generally regarded<br />
as more important. High infrared transmittance can be ensured by<br />
appropriate manufacturing controls, but only at the sacrifice of<br />
ultraviolet transmittance.<br />
Visible spectrum batch-to-batch transmittance variations in<br />
synthetic fused silica are insignificant. The high ultraviolet internal<br />
transmittance of UV-grade synthetic fused silica is correlated with<br />
a visible internal transmittance that is so high it is beyond traditional<br />
methods of measurement. It is necessary to measure optical signal<br />
attenuation in fibers drawn of the material.<br />
Figure 4.2 shows a semilogarithmic comparison of the internal<br />
transmittances of UV-grade synthetic fused silica and BK7 glass.<br />
It is evident from this graph that UV-grade synthetic fused silica<br />
averages about two orders of magnitude less absorption loss than<br />
BK7 across the visible spectrum. In a sample thickness of 10 mm,<br />
the internal transmittance of UV-grade synthetic fused silica differs<br />
from unity only in the fifth decimal place. The high internal transmittance<br />
of such a material can be exploited by maintaining the optic<br />
at Brewster’s angle for the appropriate linear polarization, or with<br />
the assistance of high-efficiency antireflection coatings such as<br />
HEBBAR or one of the laser line V-coats. With these coatings it<br />
is possible to achieve external transmittances of 98.5% and 99.5%,<br />
respectively. Synthetic fused silica and HEBBAR are especially<br />
well suited to each other in visible spectrum applications.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.11
Fundamental Optics<br />
100<br />
a) LOWER LIMITS<br />
90<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
PERCENT EXTERNAL TRANSMITTANCE<br />
PERCENT EXTERNAL TRANSMITTANCE<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500<br />
WAVELENGTH IN NANOMETERS<br />
b) UPPER LIMITS<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
UVGSFS<br />
OQSFS<br />
BK7<br />
BK7<br />
UVGSFS<br />
OQSFS<br />
10<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 4.1<br />
0<br />
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0<br />
WAVELENGTH IN MICROMETERS<br />
Comparison of uncoated external transmittances for UVGSFS, OQSFS, and BK7, all 10 mm in thickness<br />
4.12 1 Visit Us OnLine! www.mellesgriot.com
PERCENT INTERNAL TRANSMITTANCE<br />
99.997<br />
99.995<br />
99.993<br />
99.991<br />
99.990<br />
99.97<br />
99.95<br />
99.93<br />
99.91<br />
99.90<br />
99.7<br />
99.5<br />
99.3<br />
99.1<br />
99.0<br />
97<br />
95<br />
93<br />
WAVELENGTH IN NANOMETERS<br />
200 400 600 800 1000 1200 1400<br />
UVGSFS<br />
BK7<br />
OH bond<br />
resonance<br />
200 400 600 800 1000 1200 1400<br />
WAVELENGTH IN NANOMETERS<br />
Figure 4.2 Semilogarithmic comparison of internal<br />
transmittances of UVGSFS and BK7<br />
The internal transmittance of UV-grade synthetic fused silica<br />
shows a pronounced dip at 950 nm, while the data for BK7 give<br />
no hint of a corresponding feature. It should be understood that BK7<br />
and UVGSFS are manufactured by very different processes. One<br />
of the many differences in these materials is that UVGSFS has a<br />
much higher content of OH chemical bonds (hydroxyl content)<br />
than does BK7. The dip in UVGSFS transmittance corresponds to<br />
the OH bond resonance.<br />
1 Malitson, I.H. “Interspecimen Comparison of the Refractive Index<br />
of Fused Silica,” Journal of the <strong>Optical</strong> Society of America 55, no. 10<br />
(October 1965): 1205–1209.<br />
SYNTHETIC FUSED-SILICA CONSTANTS<br />
Abbé Constant: 67.880.5<br />
Change of Refractive Index with Temperature (0º to 700ºC):<br />
1.28 ! 10 45 /ºC<br />
Homogeneity (maximum index variation over 10-cm aperture):<br />
2 ! 10 45<br />
Density (at 25ºC): 2.20 g/cc<br />
Continuous Operating Temperature: Maximum 900ºC<br />
Coefficient of Thermal Expansion: 5.5 ! 10 47 /ºC<br />
Specific Heat (25ºC): 0.177 cal/gºC<br />
Dispersion Formula 1 at 20ºC (l in mm):<br />
2<br />
2<br />
0.6961663l<br />
0.4079426l<br />
2<br />
n 4 1 =<br />
+<br />
2 2 2 2<br />
l 4 (0.0684043) l 4 (0.1162414)<br />
2<br />
0.8974794l<br />
+<br />
. (4.7)<br />
2 2<br />
l 4 (9.896161)<br />
Refractive Index of UV-Grade Synthetic Fused Silica*<br />
Wavelength<br />
(nm)<br />
180.0<br />
190.0<br />
200.0<br />
213.9<br />
226.7<br />
230.2<br />
239.9<br />
248.3<br />
265.2<br />
275.3<br />
280.3<br />
289.4<br />
296.7<br />
302.2<br />
330.3<br />
340.4<br />
351.1<br />
361.1<br />
365.0<br />
404.7<br />
435.8<br />
441.6<br />
457.9<br />
476.5<br />
486.1<br />
488.0<br />
496.5<br />
514.5<br />
*Accuracy 83!10 45 .<br />
Index of<br />
Refraction<br />
1.58529<br />
1.56572<br />
1.55051<br />
1.53431<br />
1.52275<br />
1.52008<br />
1.51337<br />
1.50840<br />
1.50003<br />
1.49591<br />
1.49404<br />
1.49099<br />
1.48873<br />
1.48719<br />
1.48054<br />
1.47858<br />
1.47671<br />
1.47513<br />
1.47454<br />
1.46962<br />
1.46669<br />
1.46622<br />
1.46498<br />
1.46372<br />
1.46313<br />
1.46301<br />
1.46252<br />
1.46156<br />
Wavelength<br />
(nm)<br />
532.0<br />
546.1<br />
587.6<br />
589.3<br />
632.8<br />
643.8<br />
656.3<br />
694.3<br />
706.5<br />
786.0<br />
820.0<br />
830.0<br />
852.1<br />
904.0<br />
1014.0<br />
1064.0<br />
1100.0<br />
1200.0<br />
1300.0<br />
1400.0<br />
1500.0<br />
1550.0<br />
1660.0<br />
1700.0<br />
1800.0<br />
1900.0<br />
2000.0<br />
2100.0<br />
Index of<br />
Refraction<br />
1.46071<br />
1.46008<br />
1.45846<br />
1.45840<br />
1.45702<br />
1.45670<br />
1.45637<br />
1.45542<br />
1.45515<br />
1.45356<br />
1.45298<br />
1.45282<br />
1.45247<br />
1.45170<br />
1.45024<br />
1.44963<br />
1.44920<br />
1.44805<br />
1.44692<br />
1.44578<br />
1.44462<br />
1.44402<br />
1.44267<br />
1.44217<br />
1.44087<br />
1.43951<br />
1.43809<br />
1.43659<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.13
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> Crown Glass<br />
In optical crown glass, a low-index commercial-grade glass, the<br />
index of refraction, transmittance, and homogeneity are not<br />
controlled as carefully as they are in optical-grade glasses such as<br />
BK7. <strong>Optical</strong> crown is suitable for applications in which component<br />
tolerances are fairly loose and as a substrate material for mirrors.<br />
Transmittance characteristics for optical crown are shown in<br />
figure 4.3. Relevant properties of optical crown are shown in the<br />
accompanying table.<br />
OPTICAL CROWN GLASS CONSTANTS<br />
Glass Type Designation: B270<br />
Abbé Constant:<br />
v d = 58.5<br />
Dispersion: (n F 4 n C ) = 0.0089<br />
Density: 2.55 g cm 43 at 23°C<br />
Young’s Modulus: 71.5 kN/mm 2<br />
Specific Heat: C p (20º to 100°C) = 0.184 cal/g°C<br />
Coefficient of Linear Expansion (20º to 300°C):<br />
93.3 ! 10 47 /°C<br />
Transformation Temperature: 521°C<br />
Softening Point: 708°C<br />
PERCENT EXTERNAL TRANSMITTANCE<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
Refractive Index of <strong>Optical</strong> Crown Glass<br />
Wavelength<br />
(nm)<br />
Refractive<br />
Index, n<br />
Fraunhofer<br />
Designation Source<br />
Spectral<br />
Region<br />
435.8<br />
480.0<br />
486.1<br />
546.1<br />
587.6<br />
589.0<br />
643.8<br />
656.3<br />
1.53394<br />
1.52960<br />
1.52908<br />
1.52501<br />
1.52288<br />
1.52280<br />
1.52059<br />
1.52015<br />
g<br />
F′<br />
F<br />
e<br />
d<br />
D<br />
C′<br />
C<br />
Hg arc<br />
Cd arc<br />
H 2 arc<br />
Hg arc<br />
He arc<br />
Na arc<br />
Cd arc<br />
H 2 arc<br />
Blue<br />
Blue<br />
Blue<br />
Green<br />
Yellow<br />
Yellow<br />
Red<br />
Red<br />
Transmission Values for 6-mm-thick Sample<br />
300 nm = 0.3%<br />
310 nm = 7.5%<br />
320 nm = 30.7%<br />
330 nm = 56.6%<br />
340 nm = 73.6%<br />
350 nm = 83.1%<br />
360 nm = 87.2%<br />
380 nm = 88.8%<br />
400 nm = 90.6%<br />
450 nm = 90.9%<br />
500 nm = 91.4%<br />
600 nm = 91.5%<br />
Note: Transmission in visible region (including reflection loss) = 91.7% (t = 2 mm).<br />
300 400 500<br />
1000 2000 3000<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Figure 4.3<br />
WAVELENGTH IN NANOMETERS<br />
External transmittance for 10-mm-thick uncoated optical crown glass<br />
4.14 1 Visit Us OnLine! www.mellesgriot.com
Low-Expansion Borosilicate Glass<br />
The most well-known low-expansion borosilicate glass (LEBG)<br />
is Pyrex ® made by Corning. It is well suited for applications in<br />
which high temperature, thermal shock, or resistance to chemical<br />
attack are primary considerations. On the other hand, LEBG is<br />
typically less homogeneous and contains more striae and bubbles<br />
than optical glasses such as BK7. This material is ideally suited<br />
to such tasks as mirror substrates, condenser lenses for high-power<br />
illumination systems, or windows in high-temperature<br />
environments. Because of its low cost and excellent thermal stability,<br />
it is the standard material used in test plates and optical flats. As<br />
seen in figure 4.4, transmission of LEBG extends into the<br />
ultraviolet and well into the infrared. The index of refraction in<br />
this material varies considerably from batch to batch. Typical<br />
values are shown in the accompanying table.<br />
LOW-EXPANSION BOROSILICATE GLASS CONSTANTS<br />
Abbé Constant: v d = 66<br />
Density: 2.23 g cm 43 at 25°C<br />
Young’s Modulus: 5.98 !10 9 dynes/mm 2<br />
Poisson’s Ratio: 0.20<br />
Specific Heat at 25ºC: 0.17 cal/g°C<br />
Coefficient of Linear Expansion (0° to 300°C):<br />
3.25!10 46 /°C<br />
Softening Point: 820°C<br />
Melting Point: 1250°C<br />
Pyrex ® is a registered trademark of Corning, Inc.<br />
PERCENT EXTERNAL TRANSMITTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
.2 .4 .6 .8 1 1.4<br />
WAVELENGTH IN MICROMETERS<br />
2 2.4 2.8<br />
Figure 4.4 External transmittance for 8-mm-thick uncoated<br />
low-expansion borosilicate glass<br />
Refractive Index of Low-Expansion Borosilicate Glass<br />
Wavelength<br />
(nm)<br />
486.1<br />
514.5<br />
546.1<br />
587.6<br />
643.8<br />
Refractive<br />
Index, n<br />
1.479<br />
1.477<br />
1.476<br />
1.474<br />
1.472<br />
Fraunhofer<br />
Designation<br />
F<br />
e<br />
d<br />
C′<br />
Source<br />
H 2 arc<br />
Ar laser<br />
Hg arc<br />
Na arc<br />
Cd arc<br />
Spectral<br />
Region<br />
Blue<br />
Green<br />
Green<br />
Yellow<br />
Red<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.15
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Sapphire<br />
Sapphire is a superior window material in many ways. Because<br />
of its extreme surface hardness, sapphire can be scratched by only<br />
a few substances (such as diamond or boron nitride) other than<br />
itself. Chemically inert and insoluble in almost everything except at<br />
highly elevated temperatures, sapphire can be cleaned with impunity.<br />
For example, even hydrogen fluoride fails to attack sapphire at<br />
temperatures below 300ºC. Sapphire exhibits high internal<br />
transmittance all the way from 150 nm (vacuum ultraviolet) to<br />
6000 nm (middle infrared). The external transmittance of sapphire<br />
is shown in figure 4.5. Because of its great strength, sapphire windows<br />
can safely be made much thinner than windows of other glass types,<br />
and therefore are useful even at wavelengths that are very close to<br />
their transmission limits. Because of the exceptionally high thermal<br />
conductivity of sapphire, thin windows can be very effectively cooled<br />
by forced air or other methods. Conversely, sapphire windows can<br />
easily be heated to prevent condensation.<br />
Sapphire is single-crystal aluminum oxide (Al 2 O 3 ). Because of<br />
its hexagonal crystalline structure, sapphire exhibits anisotropy in<br />
many optical and physical properties. The exact characteristics of an<br />
optical component made from sapphire depend on the orientation<br />
of the optic axis or c-axis relative to the element surface. Sapphire<br />
exhibits birefringence, a difference in index of refraction in orthogonal<br />
directions. The difference in index is 0.008 between light traveling<br />
along the optic axis and light traveling perpendicular to it. Malitson 1<br />
determined a dispersion relationship for the ordinary ray in sapphire.<br />
This formula, along with the appropriate constants is shown below<br />
(l in micrometers):<br />
A1l<br />
A2l<br />
A3l<br />
n2<br />
4 1 = 4 +<br />
+<br />
2 4<br />
2 2 4<br />
2 2<br />
l l1<br />
l l2<br />
l 4 l<br />
where<br />
A 1 = 1.023798<br />
A 2 = 1.058264<br />
A 3 = 5.280792<br />
2<br />
l1<br />
= 0.00377588<br />
2<br />
l2<br />
= 0.0122544<br />
2<br />
l = 321.3616.<br />
3<br />
2<br />
The transmission of sapphire is limited primarily by losses caused<br />
by surface reflections. The high index of sapphire makes magnesium<br />
fluoride almost an ideal single-layer antireflection coating. When<br />
a single layer of magnesium fluoride is deposited on sapphire and<br />
optimized for 550 nm, total transmission of a sapphire component<br />
can be kept above 98% throughout the entire visible spectrum.<br />
1 Malitson, I.H. “Refraction and Dispersion of Synthetic Sapphire,”<br />
Journal of the <strong>Optical</strong> Society of America 525, no. 12 (Dec. 1967): 1377.<br />
2<br />
2<br />
2<br />
3<br />
(4.8)<br />
PERCENT EXTERNAL TRANSMITTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
Figure 4.5<br />
sapphire<br />
0<br />
.1 .2 .3 .5 1 1.5<br />
WAVELENGTH IN MICROMETERS<br />
3 5 8<br />
External transmittance for 1-mm-thick uncoated<br />
SAPPHIRE CONSTANTS*<br />
Density: 3.98 g cm 43 at 25ºC<br />
Young’s Modulus*: 3.7 ! 10 10 dynes/mm 2<br />
Poisson’s Ratio*: 40.02<br />
Moh Hardness: 9 (by definition)<br />
Specific Heat at 25ºC: 0.18 cal/gºC<br />
Coefficient of Linear Expansion (0º to 500ºC):<br />
7.7 ! 10 46 /ºC<br />
Softening Point: 1800ºC<br />
*Sapphire is anisotropic in many of its properties which require tensor<br />
description. These values are averages over many directions.<br />
Refractive Index of Sapphire<br />
Wavelength<br />
(nm)<br />
265.2<br />
351.1<br />
404.7<br />
488.0<br />
514.5<br />
532.0<br />
546.1<br />
632.8<br />
1550.0<br />
2000.0<br />
Refractive Index<br />
n<br />
1.8337<br />
1.7970<br />
1.7858<br />
1.7754<br />
1.7731<br />
1.7718<br />
1.7708<br />
1.7660<br />
1.7462<br />
1.7377<br />
4.16 1 Visit Us OnLine! www.mellesgriot.com
ZERODUR ®<br />
Many optical applications require a substrate material with a<br />
near-zero coefficient of thermal expansion and/or excellent thermal<br />
shock resistance. ZERODUR ® with its very small coefficient of<br />
thermal expansion at room temperature is such a material.<br />
ZERODUR, which belongs to the glass-ceramic composite<br />
class of materials, has both an amorphous (vitreous) component and<br />
a crystalline component. This Schott glass is subjected to special<br />
thermal cycling during manufacture so that approximately 75% of<br />
the vitreous material is converted to the crystalline quartz form.<br />
The crystals are typically only 50 nm in diameter, and ZERODUR<br />
appears reasonably transparent to the eye because the refractive<br />
indices of the two phases are almost identical. However, scattering<br />
at the grain boundaries precludes the use of ZERODUR for<br />
transmissive optics.<br />
Typical of amorphous substances, the vitreous phase has a<br />
positive coefficient of thermal expansion. The crystalline phase has<br />
a negative coefficient of expansion at room temperature. The overall<br />
linear thermal expansion coefficient of the combination is almost<br />
zero at useful temperatures.<br />
Figure 4.6 shows the variation of expansion coefficient with<br />
temperature for a typical sample. The actual performance varies very<br />
slightly, batch to batch, with the room temperature expansion<br />
coefficient in the range of 80.15 ! 10 46 /ºC. By design, this material<br />
exhibits a change in the sign of the coefficient near room temperature.<br />
A comparison of the thermal expansion coefficients of ZERODUR<br />
and fused silica is shown in the figure. ZERODUR, is markedly<br />
superior over a large temperature range, makes ideal mirror<br />
substrates for such stringent applications as multiple-exposure<br />
holography, holographic and general interferometry, manipulation<br />
of moderately powerful laser beams, and space-borne imaging<br />
systems.<br />
ZERODUR ® CONSTANTS<br />
Abbé Constant: v d = 66<br />
Dispersion: (n f – n c ) = 0.00967<br />
Density: 2.53 g cm 43 a 25ºC<br />
Young’s Modulus: 9.1 ! 10 9 dynes/mm 2<br />
Poisson’s Ratio: 0.24<br />
Specific Heat at 25ºC: 0.196 cal/gºC<br />
Coefficient of Linear Expansion (20º to 300ºC) :<br />
0.0580.10 ! 10 46 /ºC<br />
Maximum Temperature: 600ºC<br />
Zerodur ® is a registered trademark of Schott Glass Technologies.<br />
Refractive Index of ZERODUR ®<br />
Wavelength<br />
THERMAL EXPANSION COEFFICIENT (X 10 –6 / °K)<br />
(nm)<br />
656.3<br />
643.8<br />
587.6<br />
546.1<br />
486.1<br />
480.0<br />
435.8<br />
.8<br />
.6<br />
.4<br />
.2<br />
0<br />
–.2<br />
–.4<br />
–.6<br />
–.8<br />
–271<br />
ZERODUR<br />
fused silica<br />
–250 –150 –50 0 50 150<br />
TEMPERATURE IN DEGREES CENTIGRADE<br />
Figure 4.6 Comparison of thermal expansion coefficients<br />
of ZERODUR ® and fused silica<br />
MIRROR SUBSTRATES<br />
ZERODUR is commonly<br />
used as a substrate for<br />
l/20 mirrors with<br />
aluminum type coatings.<br />
See Chapter 9, Mirrors,<br />
for ZERODUR coated<br />
mirrors.<br />
Fraunhofer<br />
Designation<br />
C<br />
C′<br />
d<br />
e<br />
F<br />
F′<br />
g<br />
Refractive Index<br />
n<br />
1.5394<br />
1.5399<br />
1.5424<br />
1.5447<br />
1.5491<br />
1.5497<br />
1.5544<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 4.17
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Calcium Fluoride<br />
Calcium fluoride (CaF 2 ), a cubic single-crystal material, has<br />
widespread applications in the ultraviolet and infrared spectra.<br />
CaF 2 is an ideal material for use with excimer lasers. It can be<br />
manufactured into windows, lenses, prisms, and mirror substrates.<br />
CaF 2 transmits over the spectral range of about 130 nm to<br />
10 mm as shown in figure 4.7. Traditionally, it has been used primarily<br />
in the infrared, rather than in the ultraviolet. CaF 2 occurs naturally<br />
and can be mined. It is also produced synthetically using the<br />
Stockbarger method, which is a time- and energy-consuming process.<br />
Unfortunately, achieving acceptable deep ultraviolet transmission<br />
and damage resistance in CaF 2 requires much greater material<br />
purity than in the infrared, and it completely eliminates the possibility<br />
of using mined material.<br />
To meet the need for improved component lifetime and<br />
transmission at 193 nm and below, manufacturers have introduced<br />
a variety of inspection and processing methods to identify and<br />
remove various impurities at all stages of the production process,<br />
from incoming materials through crystallization. The needs for<br />
improved material homogeneity and stress birefringence have also<br />
caused producers to make alterations to the traditional Stockbarger<br />
approach. These changes allow tighter temperature control during<br />
crystal growth, as well as better regulation of vacuum and annealing<br />
process parameters.<br />
Excimer-grade CaF 2 provides the combination of deep ultraviolet<br />
transmission (for 193 nm and even 157 nm), high damage threshold,<br />
resistance to color center formation, low fluorescence, high<br />
homogeneity, and low stress birefringence characteristics required<br />
for the most demanding deep ultraviolet applications.<br />
CALCIUM FLUORIDE CONSTANTS<br />
Density: 3.18 gm cm 43 @ 25ºC<br />
Poisson Ratio: 0.26<br />
dN/dT: 410.6!10 46 /ºC<br />
Young’s Modulus: 1.75!10 7 psi<br />
Coefficient of Linear Expansion:<br />
18.9!10 46 /ºC (from 20ºC to 60ºC)<br />
Melting Point: 1360ºC<br />
Refractive Index of Calcium Fluoride<br />
Wavelength<br />
Refractive Index<br />
(mm)<br />
n<br />
0.193<br />
1.501<br />
0.248<br />
1.468<br />
0.257<br />
1.465<br />
0.266<br />
1.462<br />
0.308<br />
1.453<br />
0.355<br />
1.446<br />
0.486<br />
1.437<br />
0.587<br />
1.433<br />
0.65<br />
1.432<br />
0.7<br />
1.431<br />
1.0<br />
1.428<br />
1.5<br />
1.426<br />
2.0<br />
1.423<br />
2.5<br />
1.421<br />
3.0<br />
1.417<br />
4.0<br />
1.409<br />
5.0<br />
1.398<br />
6.0<br />
1.385<br />
7.0<br />
1.369<br />
8.0<br />
1.349<br />
WAVELENGTH IN MICROMETERS<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
PERCENT TRANSMITTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
.2 .4 .6 .8 1.0 2 4.0 10<br />
Figure 4.7<br />
External transmittance for calcium fluoride<br />
4.18 1 Visit Us OnLine! www.mellesgriot.com
<strong>Optical</strong> <strong>Coatings</strong> 5<br />
<strong>Optical</strong> <strong>Coatings</strong> 5.2<br />
OEM and Special <strong>Coatings</strong> 5.3<br />
The Reflection of Light 5.4<br />
Single-Layer Antireflection <strong>Coatings</strong> 5.8<br />
Multilayer Antireflection <strong>Coatings</strong> 5.12<br />
Thin-Film Production 5.14<br />
Single-Layer MgF 2 Antireflection <strong>Coatings</strong> 5.17<br />
HEBBAR <strong>Coatings</strong> 5.18<br />
V-<strong>Coatings</strong> 5.23<br />
High-Reflection <strong>Coatings</strong> 5.24<br />
Metallic High-Reflection <strong>Coatings</strong> 5.25<br />
Dielectric High-Reflection <strong>Coatings</strong> 5.29<br />
MAXBRIte <strong>Coatings</strong> 5.33<br />
Laser-Line MAX-R <strong>Coatings</strong> 5.35<br />
Ultrafast Coating 5.37<br />
5.1 1<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong>
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
<strong>Optical</strong> <strong>Coatings</strong><br />
A comprehensive survey of all optical components currently in<br />
use would reveal that the vast majority are made of various types<br />
of glass. This survey would also reveal that a majority of these<br />
optics are coated with thin layers of material(s) different from the<br />
substrate. The purpose of these coatings is to modify the reflection<br />
and transmission properties at the surface of the optical element.<br />
Whenever light passes from one medium into a medium of<br />
different optical properties (most notably refractive index), part of<br />
the light (between 0% and 100%) is reflected and part of the light<br />
(between 100% and 0%) is transmitted. The intensity ratio of reflected<br />
and transmitted components is primarily a function of the difference<br />
in refractive index and the angle of incidence. For many uncoated<br />
optical glasses, reflected light typically represents a few percent of<br />
incident radiation. For designs using more than a few components,<br />
losses in transmitted light level can accumulate rapidly. More<br />
important are corresponding losses in image contrast or modulation<br />
caused by weakly reflected ghost images superimposed on the desired<br />
image. Such unwanted images are often defocused beyond recognition<br />
so that contrast reduction (rather than image confusion) is their<br />
primary effect.<br />
Applications generally require that the reflected portion of<br />
incident light approach 0% for transmitting optics (lenses) and<br />
100% for reflective optics (mirrors), or is at some fixed intermediate<br />
value for partial reflectors (beamsplitters). The only applications<br />
that do not require coated optics involve transmitting optics in<br />
which only a few surfaces are in the optical path, where transmission<br />
inefficiencies may be tolerable.<br />
In principle, the surface of any optical element can be coated with<br />
thin layers of various materials (called thin films) in order to ensure<br />
the desired reflection/transmission ratio. Unfortunately, with the<br />
exception of simple metallic coatings, this ratio depends on the<br />
nature of the material from which the optic is fabricated, as well as<br />
the wavelength and angle of incidence. There is also a polarization<br />
dependence to this ratio when the angle of incidence is not 0 degrees.<br />
A multilayer coating (sometimes more than 100 individual layers)<br />
can optimize the reflection/transmission ratio for several sets of<br />
conditions (wavelength and angle of incidence) or optimize it over<br />
a particular range of conditions.<br />
Melles Griot is the leading supplier of precision simple optics.<br />
Because optics for most applications require a coating of some sort,<br />
it would not have been possible to achieve this market-leading position<br />
without our extensive knowledge of thin-film coatings. With<br />
the state-of-the-art coating department located in Irvine, California,<br />
as well as other coating facilities in Japan; Rochester, New York; and<br />
the British Isles, Melles Griot is able not only to coat large volumes<br />
of catalog and special optics, but also to develop and evaluate new<br />
coatings for special customer requirements.<br />
With new and expanded coating capabilities, Melles Griot now<br />
offers the same high-quality coatings as a separate service to<br />
customers wishing to supply their own substrates. As with any<br />
special or OEM order, please contact Melles Griot to discuss your<br />
requirements with one of our qualified applications engineers.<br />
Today, dielectric coatings are remarkably hard and durable.<br />
With proper care and handling, they can have a long life. In fact,<br />
the surface of many high-index glasses that are prone to staining can<br />
be protected with a durable antireflection coating. Several factors<br />
influence coating durability. Coating designs should be optimized<br />
for minimal overall thickness to reduce mechanical stress. The most<br />
resilient materials should be used. Great care should be taken in coating<br />
fabrication to ensure high-quality, nongranular, even layers.<br />
Although we cannot prevent accidental abuse of coated optics,<br />
Melles Griot concentrates on these other factors to produce coatings<br />
that are as durable as possible.<br />
Although the Melles Griot optical-coating departments have<br />
many years of experience in designing and fabricating various types<br />
of dielectric and metallic coatings, the science of thin films is still<br />
developing rapidly. Melles Griot monitors and incorporates new<br />
technology so that we are always able to offer the most advanced<br />
coatings available.<br />
The Melles Griot range of coatings currently includes antireflection,<br />
metallic reflectors, all-dielectric reflectors, hybrid reflectors,<br />
partial reflectors (beamsplitters), and filters for monochromatic,<br />
dichroic, or broadband applications. Many of the coatings can be<br />
applied to the simple optics described in this catalog; some coatings<br />
can be applied only to a specific range of products; and some<br />
of the coatings are supplied only as an integral part of a specific product<br />
(e.g., cube beamsplitters).<br />
If you require a special coating not described in this catalog, please<br />
contact a Melles Griot applications engineer to discuss our special<br />
coating design services.<br />
5.2 1 Visit Us OnLine! www.mellesgriot.com
OEM and Special <strong>Coatings</strong><br />
Melles Griot maintains coating capabilities at each of its lens<br />
fabrication facilities worldwide, including the Irvine, California,<br />
Photonics Components facility.<br />
In the last few years, Melles Griot has expanded and improved<br />
this coating facility to take advantage of the latest developments in<br />
thin-film technology. The resulting operation can provide highvolume<br />
coatings at competitive prices to OEM customers, as well<br />
as specialized, high-performance coatings for the most demanding<br />
user.<br />
The most important aspect of our coating capabilities is our<br />
expert design and manufacturing staff. This group blends years of<br />
practical experience with recent academic research knowledge.<br />
With a thorough understanding of both design and production<br />
issues, Melles Griot excels at producing repeatable, high-quality<br />
coatings at competitive prices.<br />
USER-SUPPLIED SUBSTRATES<br />
Melles Griot not only coats catalog and custom optics with<br />
standard and special coatings, but also applies these coatings to<br />
user-supplied substrates. A significant portion of our coating<br />
business involves applying standard or slightly modified catalog<br />
coatings to special substrates.<br />
HIGH VOLUME<br />
The high-volume output capabilities of the Melles Griot coating<br />
departments result in very competitive pricing for large-volume<br />
special orders. Even the small-order customer benefits from this<br />
large volume. Small quantities of special substrates can be coated<br />
with popular catalog coatings during routine production runs at a<br />
very modest cost.<br />
CUSTOM DESIGNS<br />
A large portion of the work carried out at Melles Griot coating<br />
facilities is special coatings designed and manufactured to customer<br />
specifications.<br />
These designs cover a wide range of wavelengths, from infrared<br />
to ultraviolet, and applications ranging from basic research through<br />
the design and manufacture of industrial and medical products. The<br />
most common special coating requests are for modified catalog<br />
coatings, which usually involve a simple shift in the design wavelength.<br />
TECHNICAL SUPPORT<br />
Melles Griot applications engineers are available to discuss your<br />
system requirements at any stage. This can make a significant<br />
difference to overall coating cost. Often a simple modification to a<br />
system design can enable catalog components or coatings to be<br />
substituted for special designs at a reduced cost, without affecting<br />
performance.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.3
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
The Reflection of Light<br />
REFLECTIONS AT UNCOATED SURFACES<br />
Whenever light is incident on the boundary between two media,<br />
some light is reflected and some is transmitted (undergoing<br />
refraction) into the second medium. Several physical laws govern<br />
the direction, phase, and relative amplitude of the reflected light.<br />
For our purposes, it is necessary to consider only polished optical<br />
surfaces. Diffuse reflections from rough surfaces are not considered<br />
here.<br />
The law of reflection states that the angle of incidence equals the<br />
angle of reflection. This is illustrated in figure 5.1 which shows<br />
reflection of a light ray at a simple air/glass interface. The incident<br />
and reflected rays make an equal angle with the axis perpendicular<br />
to the interface between the two media.<br />
INTENSITY<br />
At a simple interface between two dielectric materials, the<br />
amplitude of reflected light is a function of the ratio of the refractive<br />
index of the two materials, polarization of the incident light, and<br />
the angle of incidence.<br />
When a beam of light is incident on a plane surface at normal<br />
incidence, the relative amplitude of the reflected light, as a proportion<br />
of the incident light, is given by<br />
(14<br />
p)<br />
(1 + p)<br />
where p is the ratio of the refractive indices of the two materials<br />
(n 1 /n 2 ). Intensity is the square of this expression.<br />
air n = 1.00<br />
glass n = 1.52<br />
Figure 5.1<br />
interface<br />
incident<br />
ray<br />
reflected<br />
ray<br />
v i v r<br />
v i = v r<br />
v t<br />
refracted<br />
ray<br />
sinv t n<br />
= air<br />
sinv i n glass<br />
Reflection and refraction at a simple air/glass<br />
(5.1)<br />
The amount of reflected light is therefore larger when the<br />
disparity between the two refractive indices is greater. For an air/glass<br />
interface with the glass having a refractive index of 1.5, the intensity<br />
of the reflected light will be 4% of the incident light. For an<br />
optical system containing ten such surfaces, this shows that the<br />
transmitted beam will be attenuated to 66% of the incident beam<br />
from reflection losses alone.<br />
INCIDENCE ANGLE<br />
The intensity of reflected and transmitted beams is also a function<br />
of the angle of incidence. Because of refraction effects, it is necessary<br />
to consider internal and external reflection separately at this point.<br />
External reflection is defined as reflection at an interface where the<br />
incident beam originates in the material of lower refractive index<br />
(i.e., air in the case of an air/glass or air/water interface). Internal<br />
reflection refers to the opposite case.<br />
EXTERNAL REFLECTION AT A DIELECTRIC BOUNDARY<br />
Fresnel’s laws of reflection precisely describe amplitude and<br />
phase relationships between reflected and incident light at a<br />
boundary between two dielectric media. It is convenient to think<br />
of incident radiation as the superposition of two plane-polarized<br />
beams, one with its electric field parallel to the plane of incidence<br />
(p-polarized) and the other with its electric field perpendicular<br />
to the plane of incidence (s-polarized). Fresnel’s laws can be<br />
summarized in the following two equations which give the reflectance<br />
of the s- and p-polarized components:<br />
( )<br />
( )<br />
⎡sin<br />
v14v<br />
⎤<br />
2<br />
r s = ⎢<br />
⎥<br />
⎣⎢<br />
sin v1 + v2<br />
⎦⎥<br />
2<br />
⎡<br />
v14v2)<br />
⎤<br />
r p = ⎢<br />
⎥ .<br />
⎣⎢<br />
tan( v1 + v2)<br />
⎦⎥<br />
⎛<br />
r = n 4 1 ⎞<br />
⎜<br />
⎝ n + 1<br />
⎟<br />
⎠<br />
.<br />
2<br />
2<br />
(5.2)<br />
(5.3)<br />
In the limit of normal incidence in air, Fresnel’s laws reduce to<br />
the following simple equation:<br />
(5.4)<br />
It can easily be seen that, for a refractive index of 1.52 (crown<br />
glass), this gives a reflectance of 4%. This important result shows<br />
that about 4% of all illumination incident normal to an air-glass<br />
surface will be reflected. In a multielement lens systems, reflection<br />
losses would be very high if antireflection coatings were not used.<br />
The variation of reflectance with angle of incidence for both the<br />
s- and p-polarized components can be seen in figure 5.2. It can be<br />
seen that the reflectance remains close to 4% over about 30 degrees<br />
incidence, and that it rises rapidly to 100% at grazing incidence. In<br />
addition, note that the p-component vanishes at 56º 39′. This angle,<br />
called Brewster’s angle, is the angle at which the reflected light is<br />
5.4 1 Visit Us OnLine! www.mellesgriot.com
PERCENT REFLECTANCE<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0 10 20 30 40 50 60 70 80 90<br />
Figure 5.2 External reflection at a glass surface (n = 1.52)<br />
showing s- and p-polarized components<br />
completely polarized (see figure 5.3). This situation occurs when<br />
the reflected and refracted rays are perpendicular to each other<br />
(v 1 + v 2 = 90º ). This leads to the expression for Brewster’s angle, v B:<br />
v 1 = v B = arctan (n 2 /n 1 ).<br />
s-plane<br />
p-plane<br />
ANGLE OF INCIDENCE IN DEGREES<br />
Under these conditions, electric dipole oscillations of the p-<br />
component will be along the direction of propagation and therefore<br />
cannot contribute to the reflected ray. At Brewster’s angle, reflectance<br />
of the s-component is about 15%.<br />
INTERNAL REFLECTION AT A DIELECTRIC BOUNDARY<br />
For light incident from a higher to a lower refractive index<br />
medium, we can apply the results of Fresnel’s laws in exactly the<br />
same way. The angle in the high-index material at which polarization<br />
occurs is smaller by the ratio of the refractive indices in accordance<br />
with Snell’s law. The internal polarizing angle is 33º2l′ for<br />
a refractive index of 1.52, corresponding to the Brewster angle<br />
(56º 39′) in the external medium as shown in figure 5.4.<br />
The angle at which the emerging refracted ray is at grazing incidence<br />
is called the critical angle (see figure 5.5). For an external<br />
medium of air or vacuum (n = 1), the critical angle is given by<br />
vc ( l ) = arc sin ⎛ 1 ⎞<br />
⎜ ⎟<br />
⎝ n( l)<br />
⎠<br />
and depends on the refractive index n(l), which is a function of<br />
wavelength. For all angles of incidence higher than the critical angle,<br />
total internal reflection occurs.<br />
v p<br />
(5.5)<br />
air or vacuum<br />
index n 1<br />
p-polarized<br />
incident ray<br />
isotropic dielectric solid<br />
index n 2<br />
dipole axis<br />
direction<br />
e 1<br />
normal<br />
v 2<br />
v 1<br />
absent p-polarized<br />
reflected ray<br />
refracted ray<br />
dipole radiation<br />
pattern: sin 2 v<br />
p-polarized<br />
refracted ray<br />
Figure 5.3 Brewster’s angle (at this angle, the p-polarized<br />
component is completely absent in the reflected ray)<br />
n air<br />
n glass<br />
v c = critical angle<br />
d<br />
c<br />
b<br />
a<br />
vc<br />
Figure 5.4 Internal reflection at a glass surface (n = 1.52)<br />
showing s- and p-polarized components<br />
PERCENT REFLECTANCE<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
Brewster<br />
total reflection<br />
angle<br />
33°<br />
PRODUCT<br />
21'<br />
NUMBER A B<br />
07 PHT 501/07 PHF 501 10 3<br />
07 PHT 503/07 PHF 503 15 5<br />
07 PHT 505/07 PHF 505 20 5<br />
07 PHT 507/07 PHF 507 30 5<br />
07 PHT 509/07 PHF critical 509 angle 40 5<br />
r<br />
07 PHT s<br />
511/07 PHF 41° 5118'<br />
50 5<br />
r p<br />
0 10 20 30 40 50 60 70 80 90<br />
ANGLE OF INCIDENCE IN DEGREES<br />
Figure 5.5 Critical angle (at this angle, the emerging ray is at<br />
grazing incidence)<br />
a<br />
a<br />
b<br />
c<br />
b<br />
d<br />
c<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.5
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
PHASE CHANGES ON REFLECTION<br />
There is another, more subtle difference between internal and<br />
external reflections. During external reflection, light waves undergo<br />
a 180-degree phase shift. No such phase shift occurs for internal<br />
reflection (except in total internal reflection). This is one of the<br />
important principles on which multilayer films operate.<br />
INTERFERENCE<br />
Quantum theory shows us that light has wave/particle duality.<br />
In most classical optics experiments, it is generally the wave properties<br />
that are most important. With the exception of certain laser systems<br />
and electro-optic devices, the transmission properties of light through<br />
an optical system can be well predicted and rationalized by wave<br />
theory.<br />
One consequence of the wave properties of light is that waves<br />
exhibit interference effects. Light waves that are in phase with each<br />
other undergo constructive interference, (see figure 5.6). Light waves<br />
that are exactly out of phase with each other (by 180 degrees or<br />
p radians) undergo destructive interference, and their amplitudes<br />
cancel. In intermediate cases, total amplitude is given by the vector<br />
resultant, and intensity is given by the square of amplitude.<br />
Various experiments and instruments demonstrate light<br />
interference phenomena. Some interference effects are possible<br />
only with coherent sources (i.e., lasers), but many are produced by<br />
incoherent light. Three of the best-known demonstrations of visible<br />
light interference are Young’s slits experiment, Newton’s rings, and<br />
the Fabry-Perot interferometer. These are described in most elementary<br />
optics and physics texts.<br />
In all of these demonstrations, light from a source is split in<br />
some way to produce two similar wavefronts. The wavefronts are<br />
recombined with a variable path difference between them. Whenever<br />
the path difference is an integral number of half wavelengths (and<br />
if the wavefronts are of equal intensity), they cancel by destructive<br />
interference (i.e., an intensity minimum is produced). An intensity<br />
minimum is still produced if the interfering wavefronts are of differing<br />
amplitude; the result is just non-zero. When the path difference is<br />
an integral number of wavelengths, their intensities sum by constructive<br />
interference, and an intensity maximum is produced.<br />
AMPLITUDE AMPLITUDE<br />
constructive interference<br />
TIME<br />
destructive interference<br />
zero amplitude<br />
TIME<br />
wave I<br />
wave II<br />
resultant<br />
wave<br />
wave I<br />
wave II<br />
resultant<br />
wave<br />
Figure 5.6 A simple representation of constructive and<br />
destructive wave interference<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
THIN-FILM INTERFERENCE<br />
Thin-film coatings also rely on the principles of interference.<br />
Thin films are dielectric or metallic materials whose thickness is<br />
comparable to, or less than, the wavelength of light.<br />
When a beam of light is incident on a thin film, some of the<br />
light will be reflected at the front surface, and some of light will be<br />
reflected at the rear surface as shown in figure 5.7. The remainder<br />
will be transmitted. At this stage, we shall ignore multiple reflections.<br />
The two reflected wavefronts can interfere with each other. This<br />
5.6 1 Visit Us OnLine! www.mellesgriot.com
will depend on the ratio of optical thickness of the material and<br />
the wavelength of the incident light (see figure 5.8). The optical<br />
thickness of an element is defined as the equivalent vacuum thickness<br />
(i.e., the distance that light would travel in vacuum in the same<br />
amount of time as it takes to traverse the optical element of interest).<br />
In other words, the optical thickness of a piece of material is the<br />
thickness of that material corrected for the apparent change of<br />
wavelength passing through it.<br />
air n 0 ~1.00<br />
air n 0<br />
l<br />
n 0<br />
t = 1.5l/n = 0.75l<br />
t op = tn = 1.5l<br />
front and back<br />
surface reflections<br />
homogeneous<br />
thin<br />
film<br />
refractive<br />
index = n<br />
t<br />
physical<br />
thickness<br />
dense<br />
medium<br />
n≈2.00<br />
t op<br />
optical thickness<br />
transmitted light<br />
optical thickness<br />
of film, t op = nt<br />
Figure 5.8 A schematic diagram showing the effects of<br />
lower light velocity in a dense medium (in this example, the<br />
velocity of light is halved in the dense medium n = n/n 0 , and the<br />
optical thickness of the medium is 2!the real thickness)<br />
t<br />
l<br />
n<br />
l<br />
n 0<br />
Figure 5.7 Front and back surface reflections for a thin<br />
film at near-normal incidence<br />
The optical thickness is given by t op = t ! n, where t is the<br />
physical thickness, and n is the ratio of the speed of light in the<br />
material to the speed of light in vacuum:<br />
n = c (vacuum) .<br />
v (medium)<br />
To a very good approximation, n is the refractive index of the<br />
material.<br />
Returning to the thin film at normal incidence, the phase<br />
difference between the reflected wavefronts is given by (t op /l) !<br />
2p, where l is the wavelength of light, as usual, plus any phase<br />
differences caused by reflections at the surfaces. Clearly, if the<br />
wavelength of the incident light and the thickness of the film are<br />
such that a phase difference exists between reflections of p, then<br />
reflected wavefronts interfere destructively, and overall reflected<br />
intensity is a minimum. If the two reflections are of equal amplitude,<br />
then this amplitude (and hence intensity) minimum will be zero.<br />
In the absence of absorption or scatter, the principle of<br />
conservation of energy indicates all “lost” reflected intensity will<br />
appear as enhanced intensity in the transmitted beam. The sum<br />
of the reflected and transmitted beam intensities is always equal<br />
to the incident intensity. This important fact has been confirmed<br />
experimentally.<br />
Conversely, when the total phase shift between two reflected<br />
wavefronts is equal to zero (or multiples of 2p), then the reflected<br />
intensity will be a maximum, and the transmitted beam will be<br />
reduced accordingly.<br />
Spectrophotometer used to obtain a transmission and<br />
reflectance measurement from a optical coating<br />
(5.6)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.7
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Single-Layer Antireflection <strong>Coatings</strong><br />
The simple principles of single-layer antireflection coatings should<br />
now be clear. The substrate (glass, quartz, etc.) is coated with a thin<br />
layer of material so that reflections from the outer surface of the film<br />
and the outer surface of the substrate cancel each other by destructive<br />
interference. The intensity of the transmitted beam is correspondingly<br />
increased so that, ignoring scattering and absorption,<br />
incident energy = reflected energy + transmitted energy.<br />
Two requirements create an exact cancellation of reflected beams<br />
with a single-layer coating: The reflections are exactly 180 degrees<br />
(p radians) out of phase, and they have the same intensity,<br />
FILM THICKNESS<br />
The thickness of a single-layer antireflection film must be an odd<br />
number of quarter wavelengths in order to achieve the correct phase<br />
for cancellation. This requirement is shown in figure 5.9, which<br />
explains the mechanism of a hypothetically perfect single-layer antireflection<br />
coating. There is a p/2 phase shift for reflections at both interfaces<br />
because they are low to high index medium interfaces. These<br />
identical phase shifts cancel each other out. The net phase shift<br />
between the two reflections is therefore determined solely by the<br />
optical path difference 2t ! n c , where t is the physical thickness of<br />
the coating layer and n c is the refractive index of the coating material.<br />
The phase shift is therefore 2tn/l.<br />
Single-layer antireflection coatings are generally deposited with<br />
a thickness of l/4, where l is the desired wavelength for peak<br />
performance. The phase shift is 180 degrees (p radians), and the<br />
reflections are in a condition of exact destructive interference.<br />
air<br />
n 0<br />
thin<br />
film<br />
n<br />
glass<br />
n = 1.52<br />
wavelength<br />
= l<br />
If t op, the optical<br />
thickness (nt) = l/4,<br />
then reflections<br />
interfere destructively<br />
REFRACTIVE INDEX<br />
The intensity of a reflected beam from a single surface, at normal<br />
incidence, is given by<br />
[(1 4 p) / (1 + p)] 2 ! the incident intensity<br />
(5.7)<br />
where p is the ratio of the refractive indices of the two materials at<br />
the interface.<br />
For the two reflected beams to be equal in intensity, it is necessary<br />
that p, the refractive index ratio, be the same at both the interfaces<br />
n<br />
n<br />
air<br />
film<br />
=<br />
n<br />
n<br />
film<br />
substrate<br />
(i.e., the three refractive indices must form a geometric progression).<br />
Since the refractive index of air is 1.0, the thin antireflection<br />
film ideally should have a refractive index of √}}}}} n substrate }}}}}. <strong>Optical</strong><br />
glasses typically have refractive indices of between 1.5 and 1.75.<br />
Unfortunately, there is no ideal material that can be deposited in<br />
durable thin layers with a low enough refractive index to satisfy<br />
this requirement exactly (n = 1.23 for an antireflection coating on<br />
crown glass). However, magnesium fluoride (MgF 2 ) is a good<br />
compromise because it forms high-quality, stable films and has a<br />
reasonably low refractive index, 1.38 at a wavelength of 550 nm.<br />
Magnesium fluoride is probably the most widely used thin-film<br />
material for optical coatings. Although its performance is not<br />
outstanding, it represents a significant improvement over an uncoated<br />
surface. Typical crown glass surfaces reflect from 4% to 5% of visible<br />
light at normal incidence. A high-quality MgF 2 coating can reduce<br />
this value to 1.5%. For many applications this improvement is sufficient,<br />
and sophisticated multilayer coatings are not necessary.<br />
Such coatings work extremely well over a wide range of wavelengths<br />
and angles of incidence, despite the fact that the theoretical<br />
target of 0% reflectance is achieved by a film of quarter wavelength<br />
optical thickness only for normal incidence, and only if the refractive<br />
index of the coating material is exactly the geometric mean of the<br />
substrate and air. In fact, the single layer of quarter-wave-thickness<br />
MgF 2 coating designed for normal incidence makes its most<br />
significant contribution to the transmission of steep surfaces, where<br />
most rays are incident at large angles (see figure 5.10).<br />
WAVELENGTH DEPENDENCE<br />
(5.8)<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
t<br />
physical<br />
thickness<br />
resultant reflected<br />
intensity = zero<br />
Figure 5.9 Schematic representation of a single-layer<br />
antireflection coating<br />
As with any thin film, performance depends on the incident light<br />
wavelength for two reasons. First, at other than the design wavelength,<br />
film thickness is no longer the ideal l/4. This is taken into account by<br />
all thin-film design programs. A more subtle effect, which can be quite<br />
important, is caused by the change in refractive index of the coating<br />
and substrate with wavelength (i.e., dispersion). Only the most upto-date<br />
computer design packages, such as those used by Melles Griot,<br />
include this higher level of sophistication for multilayer coatings. For<br />
single-layer antireflection coatings, wavelength dependence of the<br />
coating performance can be evaluated from analytical expressions.<br />
5.8 1 Visit Us OnLine! www.mellesgriot.com
v = angle of incidence<br />
PERCENT REFLECTANCE<br />
(at 45° incidence)<br />
v<br />
MgF 2<br />
1 /4 wavelength optical thickness<br />
at 550 nm (n = 1.38)<br />
6<br />
5<br />
4<br />
3<br />
2<br />
1<br />
glass<br />
Figure 5.10 Performance of a normal incidence coating design for 550 nm working at 45 degrees compared with a 45<br />
degrees incidence coating working at 45 degrees.<br />
PERCENT REFLECTANCE<br />
AT 550 NANOMETERS<br />
subscripts: R s = reflectance for s-polarization<br />
subscripts: R av = reflectance for average polarization<br />
subscripts: R p = reflectance for p-polarization<br />
40<br />
35<br />
30<br />
25<br />
20<br />
15<br />
10<br />
5<br />
uncoated glass<br />
single-layer<br />
MgF 2<br />
R s = (normal incidence coating at 45°)<br />
R s = (45° incidence coating)<br />
R av = (normal incidence coating at 45°)<br />
0 20 40 60 80<br />
ANGLE OF INCIDENCE IN AIR (IN DEGREES)<br />
R av = (45° incidence coating)<br />
400 500 600 700<br />
WAVELENGTH IN NANOMETERS<br />
R p = (normal incidence coating at 45°)<br />
R p = (45° incidence coating)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.9
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
ANGLE OF INCIDENCE<br />
The irradiance reflectance of any thin-film coating varies with<br />
the angle of incidence. Two main effects lead to a complicated<br />
dependence of reflectance (hence transmission) on the angle of<br />
incidence. First, the path difference of the front and rear surface<br />
reflection from any layer is a function of angle. As the angle of incidence<br />
increases from zero (normal incidence), the optical path difference<br />
is decreased. The change in path difference results in a<br />
change of phase difference between the two interfering reflections<br />
in an identical manner to the phase change resulting from tilting a<br />
Fabry-Perot interferometer.<br />
The reflectance of any optical interface varies according to the<br />
angle of incidence as shown in figure 5.10. Thin-film performance<br />
evaluation at arbitrary angles of incidence is therefore quite complex,<br />
even for a simple one-layer antireflection coating. In short, the<br />
phase difference between the two pertinent reflections changes<br />
together with their relative amplitude.<br />
COATING FORMULAS (SINGLE LAYER)<br />
Because of the practical importance and wide usage of singlelayer<br />
coatings, especially at oblique incidence, it is valuable to have<br />
formulas from which coating reflectance curves, as functions of<br />
wavelength, angle of incidence, and polarization, can be calculated.<br />
COATING DISPERSION FORMULA<br />
The first step in evaluating performance of a single-layer antireflection<br />
coating is to calculate the refractive index of the film and<br />
substrate at the wavelength of interest. For optical purposes, a thin<br />
film may be considered to be perfectly homogeneous. The refractive<br />
index of MgF 2 , whether amorphous or crystalline, is connected to<br />
density with the Lorentz-Lorenz formula. The crystalline ordinary<br />
and extraordinary indices of refraction may be averaged for the<br />
amorphous phase.<br />
The formulas for crystalline MgF 2 are, respectively,<br />
43<br />
(3.5821) (10 )<br />
n o = 1.36957 +<br />
( l40.14925)<br />
and<br />
43<br />
(3.7415) (10 )<br />
n e = 1.381 +<br />
( l40.14947)<br />
for the ordinary and extraordinary rays, where l is the wavelength<br />
in microns.<br />
For the average of the ordinary and extraordinary indices of<br />
refraction,<br />
n = n( l) = 1 2<br />
(n o + n e).<br />
(5.9)<br />
(5.10)<br />
(5.11)<br />
The value 1.38 is the universally accepted amorphous film index<br />
for MgF 2 at a wavelength of 550 nanometers, which assumes a<br />
packing density of 100%. Real films, however, tend to be slightly<br />
porous. The refractive index of a real magnesium fluoride film is usually<br />
slightly lower than 1.38 because the packing density is rarely<br />
100% in practice. Because it is a complex function of the manufacturing<br />
process, packing density varies slightly from batch to<br />
batch. Air and water vapor can also settle in the film and affect its<br />
refractive index. For Melles Griot magnesium fluoride coatings,<br />
this will usually correspond to an effective refractive index between<br />
97% and 100% of the 1.38 theoretical value.<br />
COATED SURFACE<br />
REFLECTANCE AT NORMAL INCIDENCE<br />
Suppose that the coating is of quarterwave optical thickness for<br />
some wavelength l. Let n a denote the refractive index of the external<br />
medium at this wavelength (1.0 for air or vacuum), and let n f and n s ,<br />
respectively, denote the film and substrate indices. For normal incidence<br />
at this wavelength (as shown in figure 5.11), the single-pass irradiance<br />
reflectance of the coated surface can be shown to be<br />
⎛<br />
R = n n n 2<br />
a s4<br />
⎞<br />
f<br />
⎜<br />
2 ⎟<br />
⎝ n n + n ⎠<br />
a s f<br />
air or vacuum<br />
index n a<br />
wavelength l<br />
MgF 2<br />
antireflection<br />
coating<br />
index n f<br />
Figure 5.11 Reflectance at normal incidence<br />
2<br />
substrate<br />
index n s<br />
(5.12)<br />
regardless of the polarization state of the incident radiation. This<br />
function is shown in figure 5.12<br />
COATED SURFACE<br />
REFLECTANCE AT OBLIQUE INCIDENCE<br />
At oblique incidence, the situation is more complex. Let n 1 , n 2 , and<br />
n 3 , respectively, represent the wavelength-dependent refractive indices<br />
of the external medium (air or vacuum), coating film, and substrate<br />
as shown in figure 5.13. Assume that the coating exhibits a reflectance<br />
extremum of the first order for some wavelength l d and angle of<br />
incidence v 1d in the external medium. The coating is completely<br />
specified when v 1d and l d are known. One may then identify n 2 with<br />
the film index n f (1.38 for MgF 2 at 550 nm). The extremum is a<br />
minimum if n 2 is less than n 3 and a maximum if n 2 exceeds n 3 . The<br />
same formulas apply in either case.<br />
5.10 1 Visit Us OnLine! www.mellesgriot.com
PERCENT REFLECTANCE PER SURFACE<br />
2.0<br />
1.8<br />
1.6<br />
1.4<br />
1.2<br />
1.0<br />
.8<br />
.6<br />
.4<br />
.2<br />
Figure 5.12 Reflectance at surface of substrate with<br />
index n g when coated with a quarter wavelength of<br />
magnesium fluoride (index n=1.38)<br />
air or vacuum index n 1<br />
wavelength l 1<br />
MgF 2 antireflection<br />
coating index n 2<br />
fused silica<br />
glass or silica substrate<br />
index n 3<br />
BK7<br />
optical path difference = 2n 2 b–n 1 a<br />
v 1<br />
a<br />
b v 2<br />
v 3<br />
SF11<br />
LaSFN9<br />
1.4 1.5 1.6 1.7 1.8 1.9<br />
REFRACTIVE INDEX (n g )<br />
Figure 5.13 Reflectance at oblique incidence<br />
Corresponding to the angle of incidence v 1d is an angle of<br />
refraction in the film:<br />
v<br />
As v 1 is reduced from v 1d to zero, the reflectance extremum shifts<br />
in wavelength from l d to l n , where the subscript n denotes normal<br />
incidence.<br />
This wavelength is given by the equation<br />
l<br />
2d<br />
n<br />
⎛ v1d<br />
= arcsin sin ⎞<br />
⎜ .<br />
⎝ n ( l )<br />
⎟<br />
⎠<br />
2 d<br />
⎛ n 2 ( ln)<br />
⎞ ⎛ ld<br />
= ⎜<br />
⎝ n ( l )<br />
⎟ ⎜<br />
⎠ ⎝ cos v<br />
2 d<br />
2d<br />
⎞<br />
⎟ .<br />
⎠<br />
b<br />
h<br />
(5.13)<br />
(5.14)<br />
Corresponding to the arbitrary angle of incidence v 1 and<br />
arbitrary wavelength l 1 are angles of refraction in the coating and<br />
substrate, given by<br />
v<br />
and<br />
v<br />
2<br />
3<br />
⎛<br />
1 l1 v1<br />
= arcsin n ( ) sin ⎞<br />
⎜<br />
⎝ n ( l )<br />
⎟<br />
⎠<br />
2 1<br />
⎛<br />
1 l1 v1<br />
= arcsin n ( ) sin ⎞<br />
⎜<br />
⎝ n ( l )<br />
⎟<br />
⎠ .<br />
3 1<br />
Following are formulas for the single-interface amplitude<br />
reflectances for both the p- and s-polarizations:<br />
r =<br />
12p<br />
r =<br />
23p<br />
r =<br />
12s<br />
r =<br />
23s<br />
n cos v 4n cos v<br />
n cos v + n cos v<br />
2 1 1 2<br />
2 1 1 2<br />
n cos v<br />
n cos v<br />
4n cos v<br />
+ n cos v<br />
3 2 2 3<br />
3 2 2 3<br />
n cos v 4n cos v<br />
n cos v + n cos v<br />
1 1 2 2<br />
1 1 2 2<br />
n cos v<br />
n cos v<br />
4n cos v<br />
2 2 3 3<br />
2 2<br />
+ n cos v<br />
3 3<br />
The subscript “12p,” for example, means that the formula gives the<br />
amplitude reflectance for the p-polarization at the interface between<br />
the first and second media.<br />
The corresponding irradiance reflectances for the coated surface,<br />
accounting for both interferences and the phase differences between<br />
the reflected waves, are given by<br />
and<br />
R =<br />
p<br />
R =<br />
s<br />
2<br />
2<br />
r 12p + r 23p + 2r12pr 23p cos (2 b )<br />
2 2<br />
1 + r r + 2r r cos (2 b )<br />
2<br />
12p 23p<br />
2<br />
12p 23p<br />
r 12s + r 23s + 2r12sr 23s cos (2 b )<br />
2 2<br />
1 + r r + 2r r cos (2 b )<br />
12s 23s<br />
12s 23s<br />
where b is the phase difference (in the external medium) between<br />
waves reflected from the first and second surfaces of the coating.<br />
p<br />
b = 2 n 2 ( l1<br />
) h cos v2<br />
.<br />
l<br />
1<br />
The cosines must be in radians. The average reflectance is given by<br />
R = 1 2 (R p + R s ) .<br />
.<br />
(5.15)<br />
(5.16)<br />
(5.17)<br />
(5.18)<br />
(5.19)<br />
(5.20)<br />
(5.21)<br />
(5.22)<br />
(5.23)<br />
(5.24)<br />
With these formulas, reflectance curves can be calculated as functions<br />
of either wavelength l 1 or angle of incidence v 1 .<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.11
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Multilayer Antireflection <strong>Coatings</strong><br />
Previously, we discussed basic principles of thin-film design and<br />
operation for a simple antireflection coating of magnesium fluoride.<br />
It is useful to discuss to also discuss layer antireflection coatings in<br />
order to understand the operation of multilayer coatings. It is beyond<br />
the scope of this chapter to cover all aspects of modern thin-film<br />
design and operation; however, it is hoped that this section will provide<br />
the reader with insight into thin films that will be useful when<br />
considering system designs and specifying cost-effective coatings.<br />
Two basic types of antireflection coating have been developed<br />
that are worth examining in detail: the quarter/quarter coating and<br />
the multilayer broadband coating.<br />
THE QUARTER/QUARTER COATING<br />
This coating is used as an alternative to the single-layer<br />
antireflection coating. It was developed because of the lack of suitable<br />
materials available to improve the performance of single-layer<br />
coatings. The basic problem of a single-layer antireflection coating<br />
is that the refractive index of the coating material is too high,<br />
resulting in too strong a reflection from the first surface which cannot<br />
be completely canceled by interference of the weaker reflection<br />
from the substrate surface. In a two-layer coating, the first reflection<br />
is canceled by interference with two weaker reflections.<br />
A quarter/quarter coating consists of two layers, both of which<br />
have an optical thickness of a quarter wave at the wavelength of interest.<br />
The outer layer is made of a low-refractive-index material, and the<br />
inner layer is made of a high-refractive-index material (compared to<br />
the substrate). As figure 5.14 shows, the second and third reflections<br />
are both exactly 180 degrees out of phase with the first reflection.<br />
As with any multilayer coating, performance and design are<br />
calculated in terms of relative amplitudes and phases which are<br />
then summed to give the overall (net) amplitude of the reflected<br />
beam. The overall amplitude is then squared to give the intensity.<br />
How does one calculate the required refractive index of the inner<br />
layer? Several methodologies have been developed over the last 40<br />
to 50 years to calculate thin-film coating properties and converge<br />
on optimum designs. The whole field has been revolutionized in<br />
recent years with the availability of powerful microcomputers.<br />
Among the most sophisticated and effective programs are those<br />
developed by Professor H. A. Macleod, which are used by<br />
Melles Griot.<br />
With a two-layer quarter/quarter coating optimized for one<br />
wavelength at normal incidence, the required refractive indices<br />
can easily be calculated by hand. The formula for exact zero<br />
reflectance for such a coating is<br />
2<br />
3<br />
= n<br />
2 0<br />
2<br />
nn 1<br />
n<br />
(5.25)<br />
where n 0 is the refractive index of air (approximated as 1.0), n 3 is<br />
the refractive index of the substrate material, and n 1 and n 2 are the<br />
refractive indices of the two film materials, as indicated in figure 5.14.<br />
If the substrate is crown glass with a refractive index of 1.52 and<br />
if the first layer is the lowest possible refractive index, 1.38 (MgF 2 ),<br />
the refractive index of the high-index layer needs to be 1.70. Either<br />
beryllium oxide or magnesium oxide could be used for the inner layer,<br />
but both are soft materials and will not produce very durable coatings.<br />
Although it allows some freedom in the choice of coating<br />
materials and can give very low reflectance, the quarter/quarter<br />
coating is very restrictive in its design. In principle, it is possible to<br />
deposit two materials simultaneously to achieve layers of almost any<br />
required refractive index, but such coatings are not very practical.<br />
As a consequence, thin-film engineers have developed multilayer<br />
antireflection coatings and two-layer coating designs to allow the<br />
refractive index of each layer to be chosen.<br />
quarter/quarter antireflection coating<br />
A B C<br />
AMPLITUDE<br />
Figure 5.14<br />
coating<br />
TIME<br />
air (n 0 = 1.0)<br />
low-index layer (n 1 = 1.38)<br />
high-index layer (n 2 = 1.70)<br />
substrate (n 3 = 1.52)<br />
wavefront A<br />
wavefront B<br />
wavefront C<br />
resultant<br />
wave<br />
Interference in a typical quarter/quarter<br />
5.12 1 Visit Us OnLine! www.mellesgriot.com
Two-Layer <strong>Coatings</strong> of Arbitrary Thickness<br />
Interference is often thought of in terms of constructive or<br />
destructive interference, where the phase shift between interfering<br />
wavefronts is either 0 or 180 degrees. For two wavefronts to<br />
completely cancel, as in a single-layer antireflection coating, a phase<br />
shift of exactly 180 degrees is required. Where three or more reflecting<br />
surfaces are involved, complete cancellation can be achieved<br />
by carefully choosing arbitrary phase and relative intensities. This<br />
is the basis of a two-layer antireflection coating, where the layers are<br />
adjusted to suit the refractive index of available materials, instead<br />
of vice versa. For a given combination of materials, there are usually<br />
two combinations of layer thicknesses that will give zero reflectance<br />
at the design wavelength. These two combinations are of different<br />
overall thickness. For any type of thin-film coating, the thinnest<br />
possible overall coating is used since it will have better mechanical<br />
properties (less stress). In this case, the thinner combination is also<br />
less wavelength sensitive.<br />
Two-layer antireflection coatings are the simplest of the socalled<br />
V-coatings. The term V-coating arises from the shape of the<br />
reflectance curve as a function of wavelength, which is a skewed<br />
V shape with a reflectance minimum at the design wavelength (see<br />
figure 5.15). V-coatings are very popular, economical coatings for<br />
near monochromatic applications, such as optical systems using<br />
nontunable laser radiation (e.g., helium neon lasers at 632.8 nm).<br />
BROADBAND ANTIREFLECTION COATINGS<br />
Many optical systems (particularly imaging systems) use<br />
polychromatic (more than one wavelength) light. In order for the<br />
system to have a flat spectral response, transmitting optics are coated<br />
with a broadband or dichroic antireflection coating. The main<br />
technique used in designing antireflection coatings that are highly<br />
efficient at more than one wavelength is to use absentee layers within<br />
the coating. There are two additional techniques that can be used<br />
for shaping performance curves of high-reflectance coatings and<br />
wavelength-selective filters, but these are not applicable to antireflection<br />
coatings.<br />
Absentee Layers<br />
An absentee layer is a film of dielectric material that does not<br />
change the performance of the overall coating at one particular<br />
wavelength, usually the wavelength for which the coating is being<br />
principally optimized. This results from the fact that the coating has<br />
an optical thickness of a half wave at that wavelength. The effects<br />
of the “extra” reflections cancel out at the two interfaces since no<br />
additional phase shifts are introduced. In theory, the performance<br />
of the coating is the same at that wavelength whether the absentee<br />
layer is present or not.<br />
At other wavelengths, the absentee layer starts to have an effect,<br />
for two reasons. The ratio between physical thickness of the layer<br />
and the wavelength of light changes with wavelength. Also, dispersion<br />
of the coating material causes optical thickness to change with<br />
wavelength.<br />
Multilayer Broadband Antireflection <strong>Coatings</strong><br />
The complex, computer-design techniques used by Melles Griot<br />
for multilayer antireflection coatings are based on the simple principles<br />
of interference and phase shifts described in the preceding text.<br />
All methods consider the combined effect of various film elements.<br />
Because of the extensive properties of coherent interference, it is<br />
meaningless to consider individual layers in a multilayer coating.<br />
Each layer is influenced by the optical properties of the layer next<br />
to it. The properties of that layer are influenced by its environment.<br />
Clearly, this represents at least a complex series of matrix multiplications,<br />
where each matrix corresponds to a single layer.<br />
An important aspect that is often overlooked in simple theory<br />
is that there are multiple “reflections” in the coatings. In the previous<br />
discussions, only first-order reflections have been considered. This<br />
oversimplified approach is unable to predict correctly the behavior<br />
of multilayer coatings. Second, third, and higher order terms must<br />
be considered if real behavior is to be modeled accurately. The exact<br />
behavior of an antireflection coating is clearly dependent on the<br />
refractive index of the substrate to which it is applied. In order to<br />
simplify the task of choosing and ordering coatings for optics of different<br />
glass types, Melles Griot has listed the coatings in this catalog<br />
according to performance. Actual coatings applied by Melles Griot<br />
are adjusted for different glass types in order to achieve the specified<br />
performance.<br />
REFLECTANCE<br />
l 0<br />
WAVELENGTH<br />
Figure 5.15 Characteristic performance curve of<br />
a V-coating<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.13
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Thin-Film Production<br />
VACUUM DEPOSITION<br />
Melles Griot manufactures thin films by a process known as vacuum<br />
deposition. Uncoated substrates are placed in a large vacuum<br />
chamber capable of achieving a vacuum of at least 10 46 torr. At<br />
the bottom of the chamber is a source of the film material to be<br />
vaporized, as shown in figure 5.16. The substrates are mounted on<br />
a series of rotating carousels, arranged so that each substrate sweeps<br />
in planetary style through the same time-averaged volume in the<br />
chamber.<br />
THERMAL EVAPORATION<br />
The source of vaporized material is usually one of two types.<br />
The simpler, older type relies on resistive heating of a thin folded<br />
strip (boat) of tungsten, tantalum, or molybdenum by a high direct<br />
current. Small amounts of the coating material are loaded into the<br />
substrates<br />
thermocouple<br />
quartz lamp<br />
(heating)<br />
baseplate<br />
power<br />
supply<br />
filter<br />
detector<br />
rotation motor<br />
monitoring<br />
plate substrates<br />
shutter<br />
quartz lamp<br />
vapor<br />
E-beam gun<br />
chopper<br />
light source<br />
water<br />
cooling<br />
reflection signal<br />
optical monitor<br />
power<br />
supply<br />
vacuum<br />
system<br />
Figure 5.16 Schematic view of a typical vacuum deposition<br />
chamber<br />
boat. A high current (10–100 A) is passed through the boat, which<br />
undergoes resistive heating. The coating material is then vaporized<br />
thermally. Because the chamber is at a greatly reduced pressure,<br />
there is a very long mean free path for the free atoms or molecules,<br />
and the heavy vapor is able to reach the moving substrates at the<br />
top of the chamber. Here it condenses back to the solid state, forming<br />
a thin, uniform film.<br />
Several problems are associated with thermal evaporation. Some<br />
useful substances can react with the hot boat, which can cause<br />
impurities to be deposited with the layers, changing optical<br />
properties. In addition, many materials, particularly metal oxides,<br />
cannot be vaporized this way, because the material of the boat<br />
(tungsten, tantalum, or molybdenum) melts at a lower temperature.<br />
Instead of a layer of zirconium oxide, a layer of tungsten would<br />
be deposited on the substrate.<br />
For the more volatile materials, thermal evaporation is still often<br />
the method of choice. <strong>Coatings</strong> of excellent quality can be produced<br />
if they are deposited on a hot substrate.<br />
SOFT FILMS<br />
Until the advent of electron bombardment as a superior<br />
alternative, only materials that melted at moderate temperatures<br />
(2000ºC) could be incorporated into thin-film coatings. Unfortunately,<br />
the more volatile materials also happen to be the softer<br />
materials, which produce less resilient films. Consequently, early<br />
multilayer coatings deteriorated fairly quickly and required undue<br />
amounts of care during cleaning. More important, sophisticated<br />
designs with performance specifications at several wavelengths<br />
could not be easily produced since these designs required many<br />
individual layers, and the softness of the layers made some of these<br />
films impractical.<br />
ELECTRON BOMBARDMENT<br />
Electron bombardment has become the accepted method of<br />
choice for optical thin-film fabrication. This method is capable of<br />
vaporizing even highly involatile materials, such as titanium oxide and<br />
zirconium oxide. Using large cooled crucibles precludes reaction of<br />
the heated material with the metal of the boat or crucible.<br />
A high-flux electron gun (1 A at 10 kV) is aimed at the film<br />
material contained in a large, water-cooled, copper crucible. Intense<br />
local heating melts and vaporizes some of the coating material in<br />
the center of the crucible without causing undue heating of the<br />
crucible itself. For particularly involatile materials, the electron gun<br />
can be focused to intensify its effects.<br />
Careful control of temperature and vacuum conditions ensures<br />
that most of the vapor is in the form of atoms or molecules, as<br />
opposed to clusters. This produces a more even coating with better<br />
optical characteristics and improved longevity.<br />
5.14 1 Visit Us OnLine! www.mellesgriot.com
ION-ASSISTED BOMBARDMENT<br />
Ion-assisted bombardment is a coating technique that can offer<br />
unique benefits under certain circumstances. Ion assist during<br />
coating leads to a higher atomic or molecular packing density in the<br />
thin-film layers. This results in a higher refractive index and, most<br />
important, superior mechanical characteristics.<br />
Specifically, the lack of voids in the more efficiently packed<br />
film means that it is far less susceptible to water-vapor absorption.<br />
Water absorption by an optical coating can change the index of<br />
refraction of layers and, hence, the optical properties. Water absorption<br />
can also cause mechanical changes that can ultimately lead to<br />
failure.<br />
Ion-assisted coating can also be used for cold processing.<br />
Eliminating the need to heat parts allows cemented parts, such as<br />
achromats, to be safely coated.<br />
MONITORING AND CONTROLLING LAYER THICKNESS<br />
A chamber set up for multilayer deposition has several sources<br />
that are preloaded with various coating materials. The entire<br />
multilayer coating is deposited without opening the chamber.<br />
A source is heated, or the electron gun is turned on, until the<br />
source is stable. The shutter above the source is opened to expose<br />
the chamber to the vaporized material. When a particular layer is<br />
deposited to the correct thickness, the shutter is closed and the source<br />
is turned off. This process is repeated for the other sources.<br />
The most common method of monitoring the deposition process<br />
is optical monitoring. A monitor beam of light passes through the<br />
chamber and is incident on a blank monitor substrate. Reflected<br />
light is detected using photomultiplier and phase-sensitive detection.<br />
As each layer is deposited onto the reference blank, the intensity<br />
of reflected light from it oscillates in a pseudo sine wave (rectified).<br />
The turning points represent quarter- and half-wave thicknesses at<br />
the monitor wavelength, with intermediate thicknesses between.<br />
Deposition is automatically stopped as the reflectance of the reference<br />
surface passes through the appropriate point.<br />
SCATTERING<br />
Reflectance and transmittance are usually the most important<br />
optical properties specified for a thin film, closely followed by<br />
absorption. However, the degree of scattering caused by a coating<br />
is often the limiting factor in the ability of coated optics to perform<br />
in certain applications. Scattering is quite complex. The overall<br />
degree of scattering is determined by imperfections in layer interfaces<br />
and interference between photons of light scattered by these<br />
imperfections as shown in figure 5.17. It is also a function of the<br />
granularity of the layers. This is difficult to control as it is an inherent<br />
characteristic of the materials used. Careful modification of deposition<br />
conditions can make a considerable difference to this effect.<br />
The most notable examples of applications where scattering is<br />
critical are intracavity mirrors for low-gain lasers, such as certain<br />
helium neon laser lines and continuous-wave dye lasers.<br />
TEMPERATURE AND STRESS<br />
A major problem with thin films is caused by inherent mechanical<br />
stresses. Even with careful control of the vacuum, source<br />
temperature, and optimized positioning of the optics being coated,<br />
many thin-film materials do not deposit well on cold substrates.<br />
This is particularly true of involatile materials. Raising the substrate<br />
temperature a few hundred degrees improves the quality of these<br />
films, often making the difference between usable and useless film.<br />
The elevated temperature seems to allow freshly condensed atoms<br />
(or molecules) to undergo limited surface diffusion.<br />
Optics that have been given a multilayer thin-film coating at an<br />
elevated temperature require very slow cooling to room temperature.<br />
Thermal expansion coefficients of substrate and film materials are<br />
likely to be somewhat different. As cooling occurs, the coating<br />
contracts and produces stress in the layers. Many pairs of coating<br />
materials do not adhere particularly well to each other owing to<br />
different chemical properties and bulk packing characteristics.<br />
Temperature-induced stress and poor interlayer adhesion are<br />
the most common thickness limitations for optical thin films. Until<br />
new technologies, such as ion-assisted deposition, are developed<br />
into true production tools, stress must be reduced by minimizing<br />
overall coating thickness and by carefully controlling the production<br />
process.<br />
incident light<br />
Figure 5.17 Interface imperfections scattering light in a<br />
multilayer coating<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.15
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
INTRINSIC STRESS<br />
Even in the absence of thermal-contraction-induced stress, the<br />
layers often are not mechanically stable because of intrinsic stress<br />
from interatomic forces. The homogeneous thin film is not the<br />
preferred phase for most coating materials. In the lowest energy,<br />
natural form of the material, molecules are aligned in a crystalline<br />
symmetric fashion. This is the form in which intermolecular forces<br />
are more nearly in equilibrium.<br />
In addition to intrinsic molecular forces, intrinsic stress results<br />
from poor packing. If packing density is considerably less than<br />
100%, the intermolecular binding may be sufficiently weakened to<br />
make the layer totally unstable.<br />
PRODUCTION CONTROL<br />
Two major factors are involved in producing a coating to perform<br />
to a particular set of specifications. First, sound design techniques<br />
must be used. If design procedures cannot accurately predict the<br />
behavior of a coating, there is little chance that satisfactory coatings<br />
will be produced. Second, if the manufacturing phase is not carefully<br />
controlled, the thin-film coatings produced may perform quite<br />
differently from the computer simulation.<br />
Melles Griot uses the latest computer design programs with<br />
exhaustive iterations to ensure that the final design is optimized.<br />
Manufacturing high-quality thin films is not trivial. At Melles Griot,<br />
more effort is expended on monitoring thin-film manufacture than<br />
on any other single manufacturing procedure. Without such careful<br />
monitoring, the tedious design and optimization phase would<br />
be wasted.<br />
Great care is taken in coating production at every level. Not only<br />
are all obvious precautions taken, such as thorough precleaning and<br />
controlled cool down, but even the smallest details of the manufacturing<br />
process are carefully controlled. Our thoroughness and<br />
attention to detail ensures that the customer will always be supplied<br />
with the best design, manufactured to the highest standards.<br />
QUALITY CONTROL<br />
All batches of Melles Griot coatings are rigorously and thoroughly<br />
tested for quality. Even with the most careful production control,<br />
this is necessary to ensure that only the highest quality parts are<br />
shipped.<br />
Our inspection system meets the stringent demands of<br />
MIL-I-45208A and our spectrophotometers are calibrated to<br />
standards traceable to the National Institute of Standards and<br />
Technology (NIST). Upon request, we can provide complete<br />
environmental and photometric testing to MIL-C-675 and<br />
MIL-M-13508. All are firm assurances of dependability and<br />
accuracy.<br />
5.16 1 Visit Us OnLine! www.mellesgriot.com
Single-Layer MgF 2<br />
Antireflection <strong>Coatings</strong><br />
Magnesium fluoride (MgF 2 ) is commonly used for single-layer<br />
antireflection coatings because of its almost ideal refractive index<br />
(1.38 at 550 nm) and high durability. These coatings are optimized<br />
for 550 nm (Melles Griot coating suffix /066) and 670 nm (/067) for<br />
normal incidence, but as can be seen from the reflectance curves, in<br />
figures 5.18 and 5.19, they are extremely insensitive to wavelength<br />
and incidence angle. Many standard lenses in stock are coated with<br />
MgF 2 . Our precision optimized achromats (01 LAO series) are<br />
supplied standard with the /066 coating.<br />
Should you wish to specify a different wavelength and incidence<br />
angle, it is no problem to shift the coating design. Please bear in<br />
mind, however, that additional delivery time is needed for special<br />
coatings, and that care should be taken in selecting the quantity of<br />
items coated to maximize the efficiency of the coating run. Partially<br />
filled chambers result in higher unit prices.<br />
Single-layer antireflection coatings are routinely available for<br />
almost any angle of incidence and any wavelength between 200 nm<br />
in the ultraviolet and 1.6 mm in the infrared. To obtain such coatings,<br />
simply specify (for each surface of each part) the precise wavelength<br />
and angle of incidence for which reflectance is to be minimized. As<br />
the 1.6-mm wavelength is approached, the angle-of-incidence range<br />
becomes restricted to near-normal incidence. This is because of<br />
practical limitations on physical coating thickness. It is usually<br />
inadvisable to request a MgF 2 coating for any wavelength greater<br />
than 1.6 mm. Thicker MgF 2 coatings are possible, but they tend to<br />
exhibit crazing, poor adhesion, and significantly increased scattering.<br />
Single-layer antireflection coatings for use on very steeply curved<br />
or short-radius surfaces should be specified for an angle of incidence<br />
approximately half as large as the largest angle of incidence<br />
encountered by the surface. Depending on the specific application,<br />
determination of the best wavelength for use in a coating specification<br />
may require ray and energy tracing of the optical system in its<br />
anticipated environment.<br />
The effectiveness of MgF 2 as an antireflection coating is increased<br />
dramatically with increasing refractive index of the component<br />
material. This means that, for use on high-index materials, there is<br />
often little point in using more complex coatings. The reflectance<br />
curves shown are for MgF 2 on BK7 optical glass.<br />
SINGLE VERSUS MULTILAYER COATINGS<br />
While MgF 2 does not offer the same performance as multilayer<br />
coatings, such as HEBBAR (described on the following page), it<br />
is preferred in some circumstances. Specifically, on lenses with very<br />
steep surfaces, such as our 01 LAG series aspherics, MgF 2 will<br />
actually perform better than HEBBAR near the edge of the lens<br />
because the performance of a coating shifts with the angle of<br />
incidence. The shifted MgF 2 will never be worse than an uncoated<br />
lens, but, at very high angles, HEBBAR can actually be shifted to<br />
a region where its performance is worse than if there were no coating<br />
at all.<br />
PERCENT REFLECTANCE<br />
5<br />
4<br />
3<br />
2<br />
1<br />
normal and 45° incidence<br />
45°<br />
typical reflectance curves<br />
400 500 600 700<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.18 Single-layer MgF 2 coating /066<br />
$ The most popular and versatile antireflection coating<br />
for visible wavelengths<br />
$ Highly durable and most economical<br />
$ Optimized for 550 nm, normal incidence<br />
$ Relatively insensitive to changes in incidence angle<br />
$ Damage threshold: 13.2 J/cm 2 810%, 10-nsec pulse<br />
(1050 MW/cm 2 ) at 532 nm<br />
PERCENT REFLECTANCE<br />
Figure 5.19<br />
Wavelength<br />
Range<br />
(nm)<br />
5<br />
4<br />
3<br />
2<br />
1<br />
Single-Layer MgF 2 Antireflection Coating<br />
Normal Incidence<br />
On BK7<br />
(%)<br />
0°<br />
normal incidence<br />
Maximum Reflectance<br />
On Fused Silica<br />
(%)<br />
typical reflectance curve<br />
500 600 700 800<br />
WAVELENGTH IN NANOMETERS<br />
/067 Single-layer MgF 2 , visible/IR<br />
$ Optimized for 670 nm, normal incidence<br />
$ Useful for most visible and near-infrared diode wavelengths<br />
$ Highly durable and insensitive to angle<br />
$ Damage threshold: see /066 (similar specifications)<br />
COATING<br />
SUFFIX<br />
400–700<br />
520–820<br />
2.0<br />
2.0<br />
2.25<br />
2.25<br />
/066<br />
/067<br />
Note: To order this coating, append coating suffix to product number and specify which<br />
surfaces are to be coated.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.17
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
HEBBAR <strong>Coatings</strong><br />
Our HEBBAR (high-efficiency broadband antireflection)<br />
coatings provide a very low reflectance over a broad spectral<br />
bandwidth. These multilayer films, comprising alternate layers of<br />
various index materials, are combined to reduce overall reflectance<br />
to an extremely low level for the broad spectral range covered.<br />
These coatings exhibit a characteristic, double-minimum reflectance<br />
curve covering a range of some 300 nm in wavelength. The<br />
reflectance does not exceed 1.0% and is more typically below 0.6%<br />
over this entire range. Within a more limited spectral range on<br />
either side of the central peak, reflectance can be held well below<br />
0.4%. HEBBAR coatings are somewhat insensitive to angle of<br />
incidence. The effect of increasing the angle of incidence, however,<br />
is to shift the curve to slightly shorter wavelengths and to increase<br />
the long wavelength reflectance slightly. These coatings are extremely<br />
useful for high-numerical-aperture (low f-number) lenses or steeply<br />
curved surfaces. In these cases, incidence angle varies significantly<br />
over aperture.<br />
Six versions of HEBBAR coatings are offered (see figures 5.20<br />
through 5.25). Many of our components are carried in stock with<br />
a HEBBAR coating. The /078 covers most of the visible spectrum<br />
(415–700 nm) and is optimized for normal incidence. A 45-degreeincidence<br />
version of this coating (/079) is available. The infraredshifted<br />
/077 covers the range from 700 nm to 1100 nm. Two<br />
HEBBAR coatings, the /075 and the /076, are specifically optimized<br />
for diode laser wavelengths. The /074 is a modified HEBBAR<br />
coating intended for use in the range from 300 nm to 500 nm. Over<br />
this range, the reflectance is less than 1% and typically does not<br />
exceed 0.5%. Due to the special nature of the /074 coating, it is<br />
designed to be used only at an angle of incidence of 0±15 degrees,<br />
and it is not suitable for use on lenses with steeply curved surfaces.<br />
For these coatings, reflectance values apply to indices 1.47–1.55<br />
only. Other indices, while having their own designs, will have<br />
reflectance values approximately 20% higher for incidence angles<br />
from 0 to 15 degrees and 25% higher for incidence angles of 30<br />
degrees. The typical reflectance curves shown for the HEBBAR<br />
coatings are for BK7 substrates, except for /074 which is for fused<br />
silica.<br />
HEBBAR <strong>Coatings</strong><br />
Wavelength<br />
Range<br />
(nm)<br />
440–660<br />
415–685<br />
415–700<br />
415–680<br />
632.8<br />
632.8<br />
425–670<br />
425–670<br />
440–670<br />
440–670<br />
750–1100<br />
700–1100<br />
700–1150<br />
700–1100<br />
650–850<br />
650–850<br />
780–850<br />
780–850<br />
725–875<br />
300–500<br />
300–500<br />
Maximum<br />
Reflectance<br />
(%)<br />
0.6 Abs<br />
0.4 Avg<br />
1.0 Abs<br />
1.0 Abs<br />
0.3 Abs<br />
0.4 Abs<br />
0.6 Avg<br />
1.0 Abs<br />
0.4 Abs<br />
1.0 Abs<br />
0.6 Abs<br />
0.4 Avg<br />
1.0 Abs<br />
1.0 Abs<br />
0.6 Avg<br />
1.0 Abs<br />
0.25 Abs<br />
0.40 Abs<br />
1.0 Abs<br />
1.0 Abs<br />
0.5 Avg<br />
Angle of<br />
Incidence<br />
(degrees)<br />
0–15<br />
0–15<br />
0–15<br />
0–30<br />
0–15<br />
0–30<br />
45<br />
45<br />
30–45<br />
45–50<br />
0–15<br />
0–15<br />
0–15<br />
0–30<br />
0–15<br />
0–15<br />
0–15<br />
0–30<br />
0–30<br />
0–15<br />
0–15<br />
COATING<br />
SUFFIX<br />
/078<br />
/079<br />
/077<br />
/075<br />
/076<br />
/074<br />
Note: To order, append coating suffix to product number and specify which surfaces are<br />
to be coated.<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
5.18 1 Visit Us OnLine! www.mellesgriot.com
PERCENT REFLECTANCE<br />
5<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
45° incidence<br />
typical reflectance curves<br />
400 500 600 700<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.20 HEBBAR coating for visible /078<br />
$ Industry-standard multilayer, AR coating for 415 to 700 nm<br />
$ Excellent performance with HeNe and visible diode lasers<br />
$ Optimized for normal incidence<br />
$ R avg < 0.4%, R abs < 1.0%<br />
$ Damage threshold: 3.8 J/cm 2 810%,<br />
10-nsec pulse (230 MW/cm 2 ) at 532 nm<br />
PERCENT REFLECTANCE<br />
5<br />
4<br />
3<br />
2<br />
1<br />
45° incidence<br />
400 500 600 700<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.21 HEBBAR coating for visible /079<br />
typical reflectance curve<br />
$ Optimized for 425–670 nm at 45-degree incidence<br />
$ Perfect for plate beamsplitting applications<br />
$ R avg < 0.6%, R abs < 1.0%<br />
$ Damage threshold: see /078 (similar specifications)<br />
PERCENT REFLECTANCE<br />
PERCENT REFLECTANCE<br />
5<br />
4<br />
3<br />
2<br />
1<br />
5<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
45° incidence<br />
typical reflectance curves<br />
700 800 900 1000 1100 1200<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.22 HEBBAR coating for near-infrared /077<br />
$ Covers popular Ti:sapphire and diode laser wavelengths:<br />
750 to 1100 nm<br />
$ R avg < 0.4%, R abs < 0.6%<br />
$ Damage threshold: 6.5 J/cm 2 810%,<br />
20-nsec pulse (260 MW/cm 2 ) at 1064 nm<br />
normal incidence<br />
45° incidence<br />
typical reflectance curves<br />
500 550 600 650 700 750 800 850 900<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.23 HEBBAR coating for near-infrared and diode<br />
wavelengths /075<br />
$ Optimized for performance from 660 to 835 nm<br />
$ Versatile for use with most diode lasers from visible<br />
to near-infrared wavelengths<br />
$ R avg < 0.5%, R abs < 1.0%<br />
$ Damage threshold: see /078 (similar specifications)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.19
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
Fundamental Optics<br />
5<br />
typical reflectance curves<br />
5<br />
typical reflectance curves<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
PERCENT REFLECTANCE<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
45° incidence<br />
500 600 700 800 900 1000<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.24 HEBBAR coating for diode lasers /076<br />
$ Optimized for diode laser wavelengths, from 780 to 850 nm<br />
$ R avg < 0.25%, R abs < 0.4%<br />
$ Damage threshold: see /077 (similar specifications)<br />
LASER-INDUCED DAMAGE<br />
PERCENT REFLECTANCE<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
45° incidence<br />
300 350 400 450 500<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.25 HEBBAR coating for ultraviolet /074<br />
$ Excellent broadband coverage for 300 to 500 nm<br />
$ Covers HeCd and argon laser lines<br />
$ R abs < 1.0%<br />
$ Damage threshold: 3.2 J/cm 2 810%,<br />
10-nsec pulse (260 MW/cm 2 ) at 355 nm, on silica substrate<br />
Melles Griot conducts laser-induced damage testing of our optics at Big Sky Laser Technologies, Inc., in Bozeman, MT.<br />
Although the damage thresholds listed in this chapter do not constitute a performance guarantee, they are<br />
representative of the damage resistance of our coatings. Occasionally, in the damage threshold specifications, a<br />
reference is made to another coating because a suitable high-power laser is not available to test the coating within its<br />
design wavelength range. The damage threshold of the referenced coating should be an accurate representation of<br />
the coating in question.<br />
For each damage threshold specification, the information given is the peak fluence (energy per square centimeter),<br />
pulse width, peak irradiance (power per square centimeter), and test wavelength. The peak fluence is the total energy<br />
per pulse, the pulse width is the full width at half maximum (FWHM), and the test wavelength is the wavelength of the<br />
laser used to incur the damage. The peak irradiance is the energy of each pulse divided by the effective pulse length,<br />
which is from 12.5% to 25% longer than the pulse FWHM. All tests are performed at a repetition rate of 20 Hz for<br />
10 seconds at each test point. This is important because longer durations can cause damage at lower fluence levels,<br />
even at the same repetition rate.<br />
The damage resistance of any coating depends on substrate, wavelength, and pulse duration. Improper handling and<br />
cleaning can also reduce the damage resistance of a coating, as can the environment in which the optic is used. These<br />
damage threshold values are presented as guidelines and no warranty is implied. (See Chapter 14, High Energy Laser<br />
Optics for details on our guaranteed high-energy laser coatings.)<br />
When choosing a coating for its power-handling capabilities, some simple guidelines can make the decision process<br />
easier. First, the substrate material is very important. Higher damage thresholds can be achieved using fused silica<br />
instead of BK7. Second, consider the coating. Metal coatings have the lowest damage thresholds. Broadband dielectric<br />
coatings, such as the HEBBAR and MAXBRIte are better, but single-wavelength or laser-line coatings, such as the<br />
V and the MAX-R coatings, are better still. If even higher thresholds are needed, then high-energy laser (HEL) coatings<br />
are required, such as those listed in Chapter 14. If you have any questions or concerns regarding the damage levels<br />
involved in your applications, please contact a Melles Griot applications engineer.<br />
5.20 1 Visit Us OnLine! www.mellesgriot.com
EXTENDED-RANGE HEBBAR ANTIREFLECTION COATINGS<br />
Many optical systems require transmission of several, quite disparate<br />
wavelengths or transmission over a very broad continuum of wavelengths.<br />
Examples include systems involving two types of lasers, a laser<br />
system producing fundamental and harmonic wavelengths, a multiplewave<br />
mixing experiment, stimulated Raman experiments, or a system<br />
using one laser for action and another for alignment.<br />
In these situations, normal broadband coatings will not suffice.<br />
For such cases, Melles Griot makes available extended-range antireflection<br />
coatings. These coatings offer either a single, extended<br />
performance band or two separate high-performance regions. In the<br />
latter case, one of the low-reflectance regions can be quite narrow,<br />
since many of the applications often involve at least one laser beam.<br />
Melles Griot manufactures many such coatings for a variety of customer<br />
specifications. The following special coatings are offered as<br />
standard catalog items. The typical reflectance curves shown for<br />
the extended-range HEBBAR coatings are for BK7 substrates,<br />
except for /072 which is for fused silica.<br />
Visible/1064 nm Coating<br />
This coating, designated /083 and shown in figure 5.26 is<br />
designed for broadband antireflectance in the visible, as well as at<br />
1064 nm, the wavelength of Nd-YAG lasers. With less than 1%<br />
reflectance between 450 and 680 nm, and less than 0.25%<br />
reflectance at 1064 nm, this coating will find many uses in any<br />
system using a visible source in conjunction with low to moderate<br />
power Nd:YAG laser fundamental radiation. Its high performance<br />
is guaranteed for incidence angles up to 15 degrees. Optics<br />
with this coating are therefore best used at normal incidence and<br />
can be used for both converging and diverging beams.<br />
Diode Laser Coating<br />
This coating, designated /084 and shown in figure 5.27, is<br />
designed to operate at two popular diode laser wavelengths. It is a<br />
modified broadband coating which works well through the nearinfrared<br />
spectrum with reflection minima between 780 nm and<br />
830 nm and also at 1300 nm.<br />
Performance is guaranteed for incidence angles up to 15 degrees.<br />
This coating can therefore be used even without complete collimation<br />
of the diode radiation.<br />
Extended-Range HEBBAR Antireflection <strong>Coatings</strong><br />
Coating<br />
Visible / 1064 nm<br />
Diode Laser<br />
Extended Broadband<br />
UV Broadband<br />
Wavelength Range<br />
(nm)<br />
450–700<br />
1064<br />
780–830<br />
1300<br />
420–1100<br />
245–440<br />
Maximum Reflectance<br />
(%)<br />
1.25<br />
0.25<br />
0.5<br />
0.5<br />
1.75<br />
1.0<br />
Extended Broadband Coating<br />
This very broad coating, designated /073 and shown in figure<br />
5.28, offers good performance over the entire visible and nearinfrared<br />
spectral range. It is effective with broadband infrared<br />
sources, such as infrared LEDs, as well as systems using several,<br />
widely separated discrete laser lines.<br />
UV Broadband Coating<br />
The ultraviolet broadband antireflection coating, designated<br />
/072 and shown in figure 5.29, is designed for use on fused-silica<br />
substrates and provides less than 1% reflectance from 245 nm to<br />
440 nm. It is particularly useful with most popular excimer laser<br />
lines, as well as other ultraviolet sources, such as mercury lamps.<br />
The broad response of this coating allows it to perform well even<br />
with poorly collimated light, which can be especially advantageous<br />
when dealing with excimer laser sources.<br />
HIGH-ENERGY-LASER COATED OPTICS<br />
Optics and coatings specifically designed and<br />
manufactured to withstand laser-induced damage<br />
are recommended for high-power lasers, particularly<br />
pulsed lasers.<br />
See Chapter 14, High Energy Laser Optics, for an<br />
extensive listing of these products, together with a<br />
brief discussion of laser-induced damage.<br />
Average Reflectance<br />
(%)<br />
Fundamental Optics<br />
5<br />
typical reflectance curve<br />
5<br />
typical reflectance curve<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
PERCENT REFLECTANCE<br />
PERCENT REFLECTANCE<br />
4<br />
3<br />
2<br />
1<br />
400 600 800 1000 1100<br />
WAVELENGTH IN NANOMETERS<br />
5<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
Figure 5.26 HEBBAR coating for visible and Nd:YAG<br />
wavelengths /083<br />
$ Extremely versatile extended antireflection coating from<br />
450 to 700 nm and 1064 nm<br />
$ Ideal for Nd:YAG laser fundamental and second harmonic<br />
$ Also performs well across the visible, including HeNe and<br />
visible diode laser lines<br />
$ Performance guaranteed from 0º–15º on BK7 substrate<br />
$ R abs < 1.25% @ 450–700 nm, R abs < 0.25% @ 1064 nm<br />
$ Damage threshold: 1.3 J/cm 2 810%,<br />
10-nsec pulse (107 MW/cm 2 ) at 532 nm;<br />
5.4 J/cm 2 810%, 20-nsec pulse (220 MW/cm 2 ) at 1064 nm on<br />
silica substrate<br />
normal incidence<br />
WAVELENGTH IN NANOMETERS<br />
typical reflectance curve<br />
750 900 1150 1350 1500<br />
Figure 5.27 HEBBAR coating for IR diode lasers /084<br />
$ Extended antireflection coating: 780 to 830 nm and 1300 nm<br />
$ Can be used with relatively uncollimated diode laser beams<br />
$ R abs < 0.5% @ 780 nm to 830 and 1300 nm<br />
$ Damage threshold: see /083 (similar specifications at 1064 nm)<br />
PERCENT REFLECTANCE<br />
PERCENT REFLECTANCE<br />
4<br />
3<br />
2<br />
1<br />
400 500 600 700 800 900 1000 1100<br />
5<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.28 HEBBAR coating for visible and near-IR /073<br />
$ Extended antireflection coating for 420 to 1100 nm<br />
$ Excellent broadband coating, covering the visible and<br />
near-infrared regions<br />
$ R avg
V-<strong>Coatings</strong><br />
V-coatings are multilayer antireflection coatings that reduce the<br />
reflectance of a component to near-zero for one very specific wavelength.<br />
Our V-coatings are intended for use at normal incidence, for<br />
maximum reflectances of not more than 0.25% at their design wavelength.<br />
They are extremely sensitive to both wavelength and angle of<br />
incidence. For example, a V-coating intended for the helium neon<br />
wavelength (632.8 nm) when used at 30-degree incidence will reflect<br />
about 0.8%. At 45-degree incidence, the same coating will reflect over<br />
2.5%. If your application involves other than normal angles of<br />
incidence or high-numerical-aperture (low f-number) optics, it may<br />
be better to use a HEBBAR coating.<br />
Experience shows that the maximum reflectance typically<br />
achieved by these coatings is often closer to 0.1% than the 0.25%<br />
we specify. Using V-coatings on fused-silica optics can therefore<br />
provide exceptionally high external transmittances.<br />
V-Coating, Normal Incidence<br />
Wavelength<br />
(nm)<br />
193<br />
248<br />
266<br />
308<br />
351<br />
364<br />
442<br />
458<br />
466<br />
473<br />
476<br />
488<br />
496<br />
502<br />
514<br />
532<br />
543<br />
633<br />
670<br />
694<br />
780<br />
830<br />
850<br />
904<br />
1064<br />
1300<br />
1523<br />
1550<br />
Laser Type<br />
ArF<br />
ArF<br />
YAG 3rd harm.<br />
XeCl<br />
Ar ion<br />
Ar ion<br />
HeCd<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
YAG 2nd harm.<br />
HeNe<br />
HeNe<br />
GaAlAs<br />
Ruby<br />
GaAlAs<br />
GaAlAs<br />
GaAlAs<br />
GaAs<br />
Nd:YAG<br />
InGaAsP<br />
HeNe<br />
InGaAsP<br />
Maximum<br />
Reflectance<br />
(%)<br />
0.5<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
0.25<br />
COATING<br />
SUFFIX<br />
/101<br />
/102<br />
/103<br />
/104<br />
/105<br />
/107<br />
/111<br />
/112<br />
/113<br />
/114<br />
/115<br />
/116<br />
/117<br />
/118<br />
/119<br />
/122<br />
/121<br />
/123<br />
/128<br />
/124<br />
/163<br />
/166<br />
/167<br />
/125<br />
/126<br />
/168<br />
/169<br />
/169<br />
The typical reflectance curve illustrated in figure 5.30 is for a<br />
V-coating on BK7 optical glass. Clearly the performance of such a<br />
coating is highly dependent on the refractive index of the component<br />
material. However, we specify all coatings by performance<br />
figures and not by design. This means that we will change the design<br />
to suit the material being coated. This makes ordering coatings<br />
simple: select the specification you want to achieve, tell us what to<br />
put it on, and we do the rest. This means that your reflectance curve<br />
may vary from the one shown on this page, but the coating will<br />
meet your requirements. In addition to the standard coatings listed<br />
here, Melles Griot can supply V-coatings at any center wavelength<br />
from 193 nm to 2000 nm.<br />
These coatings are not intended for use with high-energy lasers<br />
although the Nd:YAG coatings are suitable for many mediumpower<br />
applications including diagnostics.<br />
PERCENT REFLECTANCE<br />
5<br />
4<br />
3<br />
2<br />
1<br />
normal incidence<br />
typical reflectance curve<br />
550 600<br />
650 700<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.30 Example of a V-coating for 632.8 nm /123<br />
$ Near-zero reflectance at one specific wavelength<br />
and incidence angle<br />
$ Maximum reflectance often less than 0.1%<br />
$ Standard coatings available for most laser lines<br />
$ Custom center wavelengths at specific angles of incidence<br />
available per request<br />
$ Damage threshold: 4.5 J/cm 2 810%,<br />
10-nsec pulse (361 MW/cm 2 ) at 532 nm for /122 on silica<br />
substrate; 10.6 J/cm 2 810%, 20-nsec pulse (480 MW/cm 2 )<br />
at 1064 nm for /126 on BK7 substrate<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.23
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
High-Reflection <strong>Coatings</strong><br />
Melles Griot offers a wide variety of high-reflection coatings<br />
for mirrors, beamsplitters, polarizing beamsplitters, dichroic mirrors,<br />
bandpass filters, and rejection filters. Some of these coatings are<br />
applied to optics as requested; others are offered only as an integral<br />
part of specialized optical elements.<br />
High-reflection coatings are ordered in the same way as antireflection<br />
coatings, namely by appending the three-digit coating<br />
suffix to the catalog number of the part being ordered.<br />
High-reflection coatings may be applied to the outside of a<br />
component, such as a flat piece of glass, to produce a first-surface<br />
mirror. Alternately, they may be applied to an internal surface to<br />
produce a second-surface mirror, such as a prism.<br />
High-reflection coatings can be categorized as either metallic or<br />
dielectric coatings.<br />
METALLIC COATINGS<br />
Metallic coatings are used primarily for mirrors and are not<br />
classified as thin films in the strictest sense. They do not rely on<br />
principles of interference, but rather on the optical properties of<br />
the coating material. However, metallic coatings are often overcoated<br />
with thin dielectric films to increase reflectance over a desired<br />
range of wavelengths or angles of incidence. In these cases, the<br />
metallic coating is said to be “enhanced.”<br />
Overcoating metallic coatings with a hard, single, dielectric layer<br />
of half-wave optical thickness improves abrasion and tarnish resistance<br />
but only marginally affects optical properties. Depending on<br />
the dielectric used, such overcoated metals are referred to as durable,<br />
protected, or hard coated.<br />
The main advantages of metallic coatings are broadband spectral<br />
performance, insensitivity to angle of incidence and polarization,<br />
and low cost. Their primary disadvantages are lower durability,<br />
lower reflectance, and lower damage threshold.<br />
DIELECTRIC COATINGS<br />
High-reflectance dielectric layers work on the same principles<br />
as dielectric antireflection coatings. Quarter-wave thicknesses of<br />
alternately high- and low-refractive index materials are applied to<br />
the substrate to form a dielectric multilayer as shown in figure 5.31.<br />
By choosing materials of appropriate refractive indices, the various<br />
reflected wavefronts can be made to interfere constructively in order<br />
to produce a highly efficient reflector.<br />
The peak reflectance value is dependent upon the ratio of<br />
refractive indices of the two materials, as well as the number of<br />
layer pairs. Increasing either increases the reflectance.<br />
The width of the reflectance curve (versus wavelength) is also<br />
determined by the film index ratio. The larger the ratio, the wider<br />
the high-reflectance region. In contrast to antireflection coatings,<br />
the inherent shape of a high-reflection coating can be modified in<br />
several different ways. The two most effective ways of modifying the<br />
performance curve are to use two or more stacks centered at slightly<br />
shifted design wavelengths, or to slightly perturb the layer thickness<br />
within a stack.<br />
Reflectance of such films can easily be made to exceed the highest<br />
metallic reflectances over limited wavelength intervals. Such films<br />
are effective for both s- and p-polarization components and over<br />
a wide angle-of-incidence range. At oblique incidence, reflectance<br />
is markedly reduced.<br />
Because of the materials chosen for the multilayer, durability<br />
and abrasion resistance of such films are normally superior to those<br />
of metallic films.<br />
air<br />
substrate<br />
Figure 5.31<br />
quarter-wave thickness of high-index material<br />
quarter-wave thickness of low-index material<br />
A simple quarter-wave stack<br />
5.24 1 Visit Us OnLine! www.mellesgriot.com
Metallic High-Reflection <strong>Coatings</strong><br />
We offer eight forms of standard metallic high-reflection coatings<br />
formed by vacuum deposition. These coatings, which can be used<br />
at any angle of incidence, can be applied to most optical components.<br />
Simply append the coating suffix number to the component product<br />
number (see figures 5.32 through 5.39).<br />
Metallic reflective coatings are delicate and require care during<br />
cleaning. Dielectric overcoats substantially improve abrasion resistance,<br />
but they are not impervious to abrasive cleaning techniques.<br />
Clean, dry pressurized gas can be used to blow off loose particles,<br />
then clean, deionized water, a mild detergent, and alcohol can be<br />
used. Gentle cleaning with a swab is recommended.<br />
ALUMINUM (/016)<br />
$ The most widely used metallic mirror coating<br />
$ Provides consistently high reflectance throughout the<br />
near-ultraviolet, visible, and near-infrared regions<br />
$ R avg > 90% from 400 to 1200 nm<br />
$ Damage threshold: 0.2 J/cm 2 810%,<br />
10-nsec pulse (12 MW/cm 2 ) at 532 nm;<br />
0.3 J/cm 2 810%, 20-nsec pulse (14 MW/cm 2 ) at 1064 nm<br />
Aluminum, the most widely used metal for reflecting films, offers<br />
consistently high reflectance throughout the visible, near-infrared,<br />
and near-ultraviolet regions of the spectrum. While silver exhibits<br />
slightly higher reflectance than aluminum through most of the visible<br />
spectrum, the advantage is temporary because of oxidation<br />
tarnishing. Aluminum also oxidizes, though more slowly, and its<br />
oxide is tough and corrosion resistant. Oxidation significantly<br />
reduces aluminum reflectance in the ultraviolet and causes slight scattering<br />
throughout the spectrum.<br />
PERCENT REFLECTANCE<br />
100<br />
95<br />
90<br />
85<br />
80<br />
normal incidence<br />
400 600 800 1000<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.32 Aluminum coating /016<br />
typical reflectance curve<br />
PROTECTED ALUMINUM (/011)<br />
$ The best general-purpose metallic reflector for visible to<br />
near-infrared<br />
$ Protective overcoat extends life of mirror and protects surface<br />
$ R avg > 87% from 400 to 800 nm<br />
$ Damage threshold: 0.3 J/cm 2 810%,<br />
10-nsec pulse (21 MW/cm 2 ) at 532 nm;<br />
0.5 J/cm 2 810%, 20-nsec pulse (22 MW/cm 2 ) at 1064 nm<br />
Protected aluminum is the very best general-purpose, metallic<br />
coating for use as an external reflector in the visible and nearinfrared<br />
spectra. Unless we specify otherwise or you specifically<br />
request a different coating, our mirrors are coated with protected<br />
aluminum. Protected aluminum is coated with a dielectric film of<br />
disilicon trioxide (Si 2 O 3 ) of half-wavelength optical thickness at<br />
550 nm. The protective film arrests oxidation and helps maintain<br />
high reflectance. It is durable enough to protect the aluminum coating<br />
from minor abrasions.<br />
PERCENT REFLECTANCE<br />
100<br />
95<br />
90<br />
85<br />
normal incidence<br />
80 45° incidence<br />
s-plane<br />
p-plane<br />
400 450 500 550 600 650 700 750<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.33 Protected aluminum coating /011<br />
typical reflectance curves<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.25
Fundamental Optics<br />
ENHANCED ALUMINUM (/023)<br />
ULTRAVIOLET-ENHANCED ALUMINUM (/028)<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
$ Enhanced performance in the mid-visible region<br />
$ Durability of protected aluminum<br />
$ R avg > 93% from 450 to 750 nm<br />
$ Damage threshold: 0.4 J/cm 2 810%,<br />
10-nsec pulse (33 MW/cm 2 ) at 532 nm,<br />
0.3 J/cm 2 810%, 20-nsec pulse (12 MW/cm 2 ) at 1064 nm<br />
By coating the aluminum with a multilayer dielectric film,<br />
reflectance is increased over a wide range of wavelengths. The<br />
durable enhancing multilayer produces a peak reflectance<br />
of 95% with an average across the visible spectrum of 93%.<br />
This coating is well suited for applications requiring the durability<br />
and reliability of protected aluminum, but with higher reflectance<br />
in the mid-visible regions. The reflectance of enhanced aluminum<br />
peaks between 530 nm and 580 nm and is high from 400 nm to<br />
800 nm.<br />
PERCENT REFLECTANCE<br />
100<br />
95<br />
90<br />
85<br />
80<br />
normal incidence<br />
45° incidence<br />
s-plane<br />
p-plane<br />
typical reflectance curves<br />
400 450 500 550 600 650 700 750<br />
WAVELENGTH IN NANOMETERS<br />
$ Maintains reflectance in the ultraviolet region<br />
$ Dielectric overcoat prevents oxidation and increases abrasion<br />
resistance<br />
$ R avg > 86% from 250 to 400 nm<br />
$ R avg > 85% from 400 to 700 nm<br />
$ Damage threshold: 0.07 J/cm 2 810%,<br />
10-nsec pulse (5.7 MW/cm 2 ) at 355 nm<br />
By applying a film of an ultraviolet-transmitting dielectric (usually<br />
MgF 2 ), the reflectance of pure, bare aluminum can be preserved in<br />
the ultraviolet. The dielectric layer prevents oxidation of the aluminum<br />
surface and provides abrasion resistance. While the resulting surface<br />
is not as abrasion resistant as our protected aluminum, this coating<br />
may be cleaned with care. Reflectance averages over 86% from 250<br />
to 400 nm and over 86% throughout the visible spectrum. This<br />
coating can be applied to all Melles Griot mirrors.<br />
PERCENT REFLECTANCE<br />
100<br />
90<br />
80<br />
70<br />
60<br />
normal incidence<br />
200 250 300 350 400<br />
WAVELENGTH IN NANOMETERS<br />
typical reflectance curve<br />
Figure 5.34 Enhanced aluminum coating /023 Figure 5.35 Ultraviolet-enhanced aluminum coating /028<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
5.26 1 Visit Us OnLine! www.mellesgriot.com
INTERNAL SILVER (/036)<br />
$ For internal reflection (second-surface) mirrors and prisms only<br />
$ Preferred for visible to near-infrared region<br />
$ Less polarization effects than aluminum<br />
$ R avg > 98% from 400 nm to 1200 nm<br />
$ Damage threshold: see /038 (similar specifications)<br />
Through most of the visible and near-infrared spectra, silver has<br />
higher reflectance than aluminum, at least for a short time following<br />
deposition. Rapid oxidation quickly causes unprotected silver coatings<br />
to deteriorate. In the internal silver coating, oxidation and tarnish are<br />
prevented by coating the external surface with an additional layer of<br />
either Inconel ® or copper. The Inconel or copper layers are subsequently<br />
painted to increase abrasion resistance. In this way, the high<br />
initial reflectance of silver is indefinitely preserved.<br />
Silver is frequently used in the near-infrared (the interval<br />
containing neodymium and gallium arsenide laser lines) because<br />
it avoids the small dip in reflectance exhibited by aluminum in this<br />
interval. In the near ultraviolet, silver has very low reflectance, and<br />
aluminum is a preferable choice. From the visible into the middleinfrared,<br />
silver offers the highest internal reflectance available from<br />
a metallic coating. Silver has less effect than aluminum on the<br />
polarization state in these regions of the spectrum.<br />
PERCENT REFLECTANCE<br />
100<br />
95<br />
90<br />
85<br />
80<br />
normal incidence<br />
400 450 500 550 600 650 700<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.36 Internal silver coating /036<br />
Inconel ® is a registered trademark of EM Industries, Inc.<br />
typical reflectance curve<br />
PROTECTED SILVER (/038)<br />
$ Extremely versatile mirror coating<br />
$ Excellent performance for the visible to infrared region<br />
$ R avg > 95% from 400 nm to 20 mm<br />
$ Can be used for ultrafast Ti:Sapphire laser applications<br />
$ Damage threshold: 0.9 J/cm 2 810%,<br />
10-nsec pulse (75 MW/cm 2 ) at 532 nm,<br />
1.6 J/cm 2 810%, 20-nsec pulse (73 MW/cm 2 ) at 1064 nm<br />
Melles Griot uses a proprietary coating and edge-sealing<br />
technology to offer a first-surface external protected silver coating.<br />
In recent tests, the protected silver coating has shown no broadening<br />
effect on a 52-femtosecond pulse. This information is presented<br />
as a guideline for femtosecond applications and no warranty is<br />
implied. Protected silver offers extremely broad performance, from<br />
400 nm to well into the infrared, with excellent durability.<br />
Due to the specialized tooling required to produce the protected<br />
silver coating, it is offered only on a limited range of substrates.<br />
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
400<br />
normal incidence<br />
typical reflectance curve<br />
500 600 700 800 900<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.37 Protected silver coating /038<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.27
Fundamental Optics<br />
BARE GOLD (/045)<br />
PROTECTED GOLD (/055)<br />
$ Widely used in the near, middle, and far infrared<br />
$ Effectively controls thermal radiation<br />
$ R avg > 99% from 700 nm to 20 mm<br />
$ Damage threshold: 1.1 J/cm 2 810% ,<br />
20-nsec pulse (48 MW/cm 2 ) at 1064 nm<br />
$ The performance of the durability of dielectrics<br />
$ Protective overcoat extends coating life<br />
$ R avg > 98% from 650 nm to 16 mm<br />
$ Damage threshold: 0.4 J/cm 2 810%,<br />
20-nsec pulse (17 MW/cm 2 ) at 1064 nm<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Because it combines good tarnish resistance with consistently<br />
high reflectance throughout the near, middle, and far infrared, gold<br />
is widely used in these regions. While it is possible to construct multilayer<br />
films that may surpass the reflectance of gold at specific<br />
wavelengths, the useful range of gold is unequaled. Gold is especially<br />
effective in controlling thermal radiation. Because bare gold is soft<br />
and scratches easily, bare-gold mirrors should be cleaned only by<br />
flow-washing with solvents and clean water or by blowing the surface<br />
clean with a low-pressure stream of dry air.<br />
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
400<br />
800 1200 1600 2000 2400 2800<br />
WAVELENGTH IN NANOMETERS<br />
The Melles Griot proprietary protected gold mirror coating<br />
combines the natural spectral performance of gold with the<br />
durability of hard dielectrics. Protected gold provides over 96%<br />
average reflectance from 650 to 1700 nm, and over 98% average<br />
reflectance from 2 to 16 mm. As well as the damage threshold listed<br />
above, the /055 coating was tested for laser-induced damage and was<br />
found to withstand up to 18±2 J/cm 2 with a 260-µs pulse<br />
(0.4 MW/cm 2 ) at a wavelength of 3 µm. These mirrors can be cleaned<br />
regularly using standard organic solvents, such as alcohol or acetone.<br />
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0<br />
normal incidence<br />
typical reflectance curve<br />
700 800 900 1000 1100 1200 1300 1400 1500 1600<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.38 Bare gold coating /045 Figure 5.39 Protected gold coating /055<br />
Coating Type<br />
Aluminum<br />
Protected Aluminum<br />
Enhanced Aluminum<br />
UV-Enhanced Aluminum<br />
Internal Silver<br />
Protected Silver<br />
Bare Gold<br />
Protected Gold<br />
normal incidence<br />
Metallic High-Reflection <strong>Coatings</strong><br />
To order, append coating suffix to product number.<br />
Wavelength Range<br />
(nm)<br />
400–1200<br />
400–800<br />
450–750<br />
250–400<br />
400–1200<br />
400–20,000<br />
700–20,000<br />
650–16,000<br />
typical reflectance curve<br />
Average Reflectance<br />
(%) COATING SUFFIX<br />
90<br />
87<br />
93<br />
86<br />
98<br />
95<br />
99<br />
98<br />
/016<br />
/011<br />
/023<br />
/028<br />
/036<br />
/038<br />
/045<br />
/055<br />
5.28 1 Visit Us OnLine! www.mellesgriot.com
Dielectric High-Reflection <strong>Coatings</strong><br />
QUARTER-WAVE STACK<br />
The basic building block for any coating involving high levels of<br />
reflection is the quarter-wave stack— a stack of alternate layers of<br />
high- and low-refractive-index material. Each layer in the stack<br />
ideally has an optical thickness of a quarter wave at the design<br />
wavelength. Alternate reflections are phase shifted by 180 degrees<br />
because they occur at low- to high-index interfaces (external<br />
reflections). These phase shifts are exactly canceled by the<br />
180-degree phase shifts caused by the path difference between<br />
alternate reflecting surfaces. All reflected wavefronts are therefore<br />
exactly in phase and undergo only constructive interference.<br />
If the difference in the refractive index of the materials is large,<br />
then a quarter-wave stack containing only a few layers will have a<br />
very high reflectance.<br />
PERFORMANCE CURVE<br />
The reflection versus wavelength performance curve of a single<br />
dielectric stack has a characteristic flat top inverted V shape as<br />
shown in figure 5.40. Clearly, reflectance is a maximum at the wavelength<br />
for which both the high- and low-index layers of the multilayer<br />
are exactly one-quarter-wave thick.<br />
Outside the fairly narrow region of high reflectance, the<br />
reflectance slowly reduces toward zero in an oscillatory fashion.<br />
Width and height (i.e., peak reflectance) of the high-reflectance<br />
region are functions of the refractive-index ratio of the two materials<br />
used, together with the number of layers actually included in the<br />
stack. The peak reflectance can be increased by adding more layers,<br />
or by using materials with a higher refractive index ratio. Amplitude<br />
reflectivity at a single interface is given by<br />
(14p)<br />
(1+ p)<br />
where<br />
⎛<br />
p = n ⎜<br />
⎝ n<br />
H<br />
L<br />
n S is the index of the substrate, and n H and n L are the indices of the highand<br />
low-index layers. N is the total number of layers in the stack.<br />
The width of the high-reflectance part of the curve (versus wavelength)<br />
is also determined by the film index ratio. The higher the<br />
ratio, the wider is the high-reflectance region.<br />
SCATTERING<br />
N41<br />
2<br />
⎞ nH<br />
⎟ × (5.26)<br />
⎠ n ,<br />
S<br />
The main parameters used to describe the performance of a<br />
thin film are reflectance and transmittance (plus absorptance, where<br />
applicable). Another less well-defined parameter is scattering. This<br />
is hard to define because of the inherent granular properties of the<br />
materials used in the films. Granularity causes some of the incident<br />
light to be “lost” by specular reflection. Often, it is scattering, not<br />
mechanical stress and weakness in the coating, that limits the<br />
maximum practical thickness of an optical coating.<br />
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0<br />
BROADBAND COATINGS<br />
RELATIVE WAVELENGTH<br />
Figure 5.40 Typical reflectance curve of an unmodified<br />
quarter-wave stack<br />
In contrast to antireflection coatings, the inherent shape of a<br />
high-reflectance coating can be modified in several different ways.<br />
The two most effective ways of modifying a performance curve are<br />
to use two or more stacks centered at slightly shifted design wavelengths,<br />
or to perturb the layer thicknesses within a stack.<br />
There is a subtle difference between multilayer antireflection<br />
coatings and multilayer high-reflection coatings, which allows the<br />
performance curves of the latter to be modified by using layer thicknesses<br />
designed for different wavelengths within a single coating.<br />
Consider a multilayer consisting of pairs, or stacks of layers, which<br />
are designed for different wavelengths. At any given wavelength,<br />
providing at least one of the layers is highly reflective for that wavelength,<br />
the overall coating will be highly reflective at that wavelength.<br />
Whether the other components transmit or are partially reflective<br />
at that wavelength is immaterial. Transmission of light of that wavelength<br />
will be blocked by reflection of a single component.<br />
On the other hand, in an antireflection coating, even if one of<br />
the stacks is exactly antireflective at a certain wavelength, the overall<br />
coating may still be quite reflective because of reflections by the<br />
other components (see figure 5.41).<br />
This can be summarized by an empirical rule. At any wavelength,<br />
the reflection of a multilayer coating consisting of several discrete<br />
components will be at least that of the most reflective component.<br />
Exceptions to this rule are coatings that have been designed to<br />
produce interference effects not just involving the surfaces within<br />
the two-layer or multilayer component stack, but also between the<br />
stacks themselves. Obvious examples are narrowband interference<br />
filters which are described in detail later and in Chapter 13, Filters.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.29
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
BROADBAND REFLECTION COATINGS<br />
The design procedure for a broadband reflection coating should<br />
now be apparent. Two design techniques are used. The most obvious<br />
approach is to use two quarter-wave stacks with their maximum<br />
reflectance wavelengths separated on either side of the design wavelength.<br />
This type of coating, however, tends to be too thick and<br />
often has poor scattering characteristics. The basic design is very<br />
useful for dichroic high reflectors, where the peak reflectances of<br />
two stacks are at different wavelengths.<br />
A more elegant approach to broadband dielectric coatings is<br />
to use a single modified quarter-wave stack. In this modified stack,<br />
the layers are not all of the same optical thickness. They are graded<br />
between the quarter-wave thickness for two wavelengths at either<br />
end of the intended broadband performance region. The optical<br />
thicknesses of the individual layers are usually chosen to follow a<br />
simple arithmetic or geometric progression. Using designs of this<br />
type, multilayer, broadband, high-reflectance coatings are possible<br />
with reflectances in excess of 99% over several hundred nanometers.<br />
The greatest impact of improved broadband reflector design<br />
and manufacturing technology has almost certainly been on dye laser<br />
design and applications. In many of these scanning systems, high<br />
reflectance over a large wavelength region is absolutely essential. In<br />
many non-laser instruments, all-dielectric coatings are favored over<br />
metallic coatings because of their high reflectance. Multilayer<br />
broadband coatings are available with high-reflectance regions<br />
spanning almost the entire visible spectrum. Such films are effective<br />
for both s- and p-polarization components, and over a wide range<br />
of incidence angles. At oblique incidence, reflectance is markedly<br />
reduced.<br />
Because of the materials chosen for the multilayer, durability<br />
and abrasion resistance of such films are superior to those of metallic<br />
films. Although the reflectance of dielectric coatings can easily be<br />
made to exceed the highest metallic reflectances over very large<br />
wavelength intervals, metallic coatings are still superior in terms of<br />
usable ranges of incidence angles and wavelengths for a single coating.<br />
POLARIZATION EFFECTS<br />
When light is incident on any optical surface at angles other<br />
than normal incidence, there is always a difference in the reflection/<br />
transmission behavior of s- and p-polarization components. In<br />
some instances, this difference can be made extremely small. On<br />
the other hand, it is sometimes advantageous to design a thin-film<br />
coating that maximizes this effect (e.g., thin-film polarizers).<br />
Polarization effects are not normally considered for antireflection<br />
coatings since these are nearly always used at normal incidence<br />
where the two polarization components are equivalent.<br />
High-reflectance or partially reflecting coatings are frequently<br />
used away from normal incidence, particularly at 45 degrees, for<br />
beam steering or beam splitting. Polarization effects can therefore<br />
be quite important for these types of coating.<br />
effective broadband high-reflection coating<br />
incident<br />
wavelength l 0<br />
NOTE: If at least one component is totally<br />
reflective, the coating will not transmit<br />
light at that wavelength.<br />
noneffective broadband antireflection coating<br />
incident<br />
wavelength l 0<br />
NOTE: Unless every component is totally<br />
nonreflective, some reflection losses will occur.<br />
totally reflective component for l 0<br />
partially reflective component for l 0<br />
totally nonreflective component for l 0<br />
Figure 5.41 Schematic multicomponent coatings with<br />
only one component exactly matched to the incident<br />
wavelength, l . (the high-reflection coating is successful; the<br />
antireflection coating is not).<br />
At certain wavelengths, a multilayer dielectric coating shows a<br />
remarkable difference in its reflectance of the s- and p-polarization<br />
components (see figure 5.42).<br />
The basis for the effect is the difference in effective refractive<br />
index of the layers of film for s- and p-components of the incident<br />
beam, as the angle of incidence is increased from zero. This should<br />
not be confused with the phenomenon of birefringence in certain<br />
crystalline materials, most notably calcite. Unlike birefringence, it<br />
does not require the symmetric properties of a truly crystalline<br />
phase. It arises from the difference in magnetic and electric field<br />
asymmetries for s- and p-components of an electromagnetic wave<br />
at oblique incidence.<br />
The resultant difference in reflectance of the two polarization<br />
components is always in the same sense. Maximum s-polarization<br />
reflectance is always greater than the maximum p-polarization<br />
5.30 1 Visit Us OnLine! www.mellesgriot.com
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
p-plane<br />
0.8 0.9 1.0 1.1 1.2<br />
RELATIVE WAVELENGTH<br />
reflectance at oblique incidence. If the reflectance is plotted as a<br />
function of wavelength for some arbitrary incidence angle, the<br />
s-polarization high-reflectance peak always extends over a broader<br />
wavelength region than the p-polarization peak.<br />
Many dielectric coatings are used at peak reflectance wavelengths<br />
where polarization differences can be made negligible. In some<br />
cases, the polarization differences can be be put to good use. The<br />
“edge” region of the reflectance curve is a wavelength region in<br />
which the s-polarization reflectance is much higher than the<br />
p-polarization reflectance. This can be maximized in a design to<br />
produce a very efficient thin-film polarizer.<br />
EDGE FILTERS AND HOT OR COLD MIRRORS<br />
s-plane<br />
Figure 5.42 The s-polarization reflectance curve is always<br />
broader and higher than the p-polarization reflectance<br />
curve<br />
In many optical systems, it is necessary to have a wavelength<br />
filtering system that transmits all light of wavelengths longer than<br />
a reference wavelength or transmits light at shorter wavelengths<br />
than a reference wavelength. These types of filters are often called<br />
short-wavelength or long-wavelength cutoff filters.<br />
Traditionally, such filters were made from colored glasses.<br />
Melles Griot offers a range of these economical and useful filters in<br />
Chapter 13, Filters. Although they are adequate for many applications,<br />
they have two drawbacks: they function by absorbing unwanted<br />
wavelengths, which may be a problem in such high-power situations<br />
as projection optics, and the edge of the transmission curve may not<br />
be as sharp as necessary for some applications.<br />
Thin films acting as edge filters are now routinely manufactured<br />
using a modified quarter-wave stack as the basic building block.<br />
Melles Griot produces many edge filters specially designed to meet<br />
customers’ specifications. A selection suitable for various laser<br />
applications is offered as standard catalog items.<br />
This type of filter is used in high-power image-projection systems<br />
where the light source often generates intense amounts of heat<br />
(infrared and near-infrared radiation). Thin-film filters designed<br />
to separate visible and infrared radiation are known as hot or cold<br />
mirrors, depending on which wavelength region is rejected (reflected)<br />
and which is transmitted. Melles Griot offers both hot and cold<br />
mirrors as catalog items (see Chapter 13, Filters).<br />
INTERFERENCE FILTERS<br />
In many applications, particularly those in the field of resonance<br />
atomic or molecular spectroscopy, a filtering system is required<br />
that transmits only a very narrow range of wavelengths of<br />
incident light. For particularly high-resolution applications, monochromators<br />
may be used, but these have very poor throughputs. In<br />
instances where moderate resolution is required and where the<br />
desired region(s) is fixed, interference filters should be used.<br />
Interference filters are produced by applying a complex multilayer<br />
coating to a colored glass blank. The complex coating consists<br />
of a series of broadband quarter-wave stacks which act as a verythin<br />
multiple-cavity Fabry-Perot interferometer. The colored glass<br />
absorbs light that would be transmitted by higher order interferences.<br />
Figure 5.43 shows the transmission curve of a typical<br />
Melles Griot interference filter, the 550-nm filter from the visible-<br />
40 filter set (03 IFS 008). Notice the square shape of the transmission<br />
curve which dies away very quickly outside the high-transmission<br />
(low-reflectance) region.<br />
More information concerning the design and operation of such<br />
filters can be found together with product listings in Chapter 13,<br />
Filters.<br />
PERCENT TRANSMITTANCE<br />
100<br />
90<br />
80<br />
70<br />
60<br />
50<br />
40<br />
30<br />
20<br />
10<br />
Figure 5.43<br />
typical transmittance curve<br />
450 550 650 750<br />
WAVELENGTH IN NANOMETERS<br />
Spectral performance of an interference filter<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.31
Fundamental Optics<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
PARTIALLY TRANSMITTING COATINGS<br />
In many applications, it is desirable to split a beam of light into<br />
two components with an arbitrary intensity ratio. This is performed<br />
by inserting an optical surface at some oblique angle (usually 45<br />
degrees) to produce a separate reflected and transmitted component.<br />
In most cases, a multilayer coating is applied to the surface in order<br />
to modify intensity and polarization ratios of the two beams.<br />
An alternative to the outdated metallic beamsplitter is a broadband<br />
(or narrowband) multilayer dielectric stack with a limited<br />
number of pairs of layers, which transmits a fixed amount of the incident<br />
light. Just as in the case of metallic beamsplitter coatings, the<br />
ratio of reflected and transmitted beams depends on the angle of<br />
incidence. Since the angle of incidence is normally fixed at 45 degrees,<br />
this does not present a significant problem. Unlike a metallic coating,<br />
a high-quality film will introduce negligible losses by either<br />
absorption or scattering. There are, however, two drawbacks to<br />
dielectric beamsplitters. The performance of these coatings is more<br />
wavelength sensitive than that of metallics, and the ratio of transmitted<br />
and reflected intensities may be quite different for the s- and<br />
p-polarization components of the incident beam. In polarizers, this<br />
can be used to advantage. The difference in partial polarization of<br />
the reflected and transmitted beams is not important, particularly<br />
where polarized lasers are used. In beamsplitters, this is usually a<br />
drawback. A hybrid metal-dielectric coating is often the best<br />
compromise.<br />
Melles Griot produces coated beamsplitters with designs ranging<br />
from broadband performance without polarization compensation,<br />
to broadband with some compensation for polarization, to a<br />
completely new range of cube beamsplitters that are virtually nonpolarizing<br />
at certain laser wavelengths. These nonpolarizing<br />
beamsplitters offer unparalleled performance with the reflected<br />
s- and p-components matched to better than 5%.<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
5.32 1 Visit Us OnLine! www.mellesgriot.com
MAXBRIte <strong>Coatings</strong><br />
MAXBRIte (multilayer all-dielectric xerophilous broadband<br />
reflecting interference) coatings are, without a doubt, the best broadband<br />
mirror coatings commercially available. The /001 coating covers<br />
the visible spectrum from 480 nm to 700 nm, the /003 is useful<br />
for diode laser applications from 630 nm to 850 nm, and the /009<br />
coating offers enhanced blue response. They all reflect well over<br />
98% of incident laser radiation within their respective wavelength<br />
ranges.<br />
These coatings exhibit exceptionally high reflectances for both<br />
s- and p-polarizations. In each case, at the most important laser<br />
wavelengths and for angles of incidence as high as 45 degrees, the<br />
average of s- and p-reflectances exceeds 99%. For most applications,<br />
they are superior to metallic or enhanced metallic coatings.<br />
The /001 MAXBRIte coating (figure 5.44) is suitable for<br />
instrumental and external laser-beam manipulation tasks. It is the<br />
ideal choice for use with tunable dye and parametric oscillator<br />
systems. The structural design of this coating is such that flatness<br />
specifications as tight as l/10 can be maintained using low-expansion<br />
substrate materials.<br />
The /003 MAXBRIte coating (figure 5.45) covers all the<br />
important diode laser wavelengths from 630 to 850 nm; therefore,<br />
it can be used with both visible and near-infrared diode lasers. This<br />
broadband coating is ideal for applications employing nontemperature-<br />
stabilized diode lasers where wavelength drift is likely to<br />
occur. The /003 also makes it possible to use a HeNe laser to align<br />
diode systems.<br />
The /007 ultraviolet MAXBRIte coating (figure 5.46) offers<br />
superior performance for ultraviolet applications. It is ideal for use<br />
with many of the excimer lasers, as well as the third and fourth harmonics<br />
of most solid-state lasers. It is also particularly useful with<br />
broadband ultraviolet light sources, such as mercury and xenon<br />
lamps. Due to mechanical stresses within this intricate coating, it<br />
is limited to substrates of l/4 figure or less.<br />
The /009 extended MAXBRIte coating (figure 5.47) offers superior<br />
response below 500 nm, and it is particularly useful for helium<br />
cadmium lasers at 442 nm, or the blue lines of argon-ion lasers. Like<br />
the /007, mechanical stresses in this complex coating limit its use to<br />
substrates of l/4 figure or less.<br />
Because of the many applications for these superior coatings we<br />
stock a number of precoated substrates.<br />
MAXBRIte <strong>Coatings</strong><br />
Wavelength<br />
Range<br />
(nm)<br />
480–700<br />
630–850<br />
245–390<br />
420–700<br />
Average<br />
Reflectance<br />
(%)<br />
98<br />
98<br />
98<br />
98<br />
Angle of<br />
Incidence<br />
(degrees)<br />
0±45<br />
0±45<br />
0±45<br />
0±45<br />
Note: To order, append coating suffix to product number.<br />
COATING<br />
SUFFIX<br />
/001<br />
/003<br />
/007<br />
/009<br />
PERCENT REFLECTANCE<br />
PERCENT REFLECTANCE<br />
100<br />
99<br />
98<br />
97<br />
96<br />
100<br />
99<br />
98<br />
97<br />
96<br />
normal incidence<br />
45° incidence<br />
500 600 700 800<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.44 MAXBRIte /001 coating<br />
normal incidence<br />
45° incidence<br />
600 700 800 900<br />
WAVELENGTH IN NANOMETERS<br />
typical reflectance curves<br />
$ Exceptional reflectance for s- and p- polarization<br />
$ Excellent performance for common visible lines<br />
$ Suitable for external laser-beam manipulation and many<br />
instrument applications<br />
$ R avg > 98% from 480 to 700 nm<br />
$ Damage threshold: 0.92 J/cm 2 810%,<br />
10-nsec pulse (57 MW/cm 2 ) at 532 nm<br />
Figure 5.45 MAXBRIte /003 coating<br />
typical reflectance curves<br />
$ Useful with visible, near-infrared diode, and HeNe lasers<br />
$ Easily accommodates diode wavelength drift<br />
$ Ideal for pointing and alignment applications<br />
$ R avg > 98% from 630 to 850 nm<br />
$ Damage threshold: see /001 (similar specifications)<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.33
<strong>Optical</strong> <strong>Coatings</strong><br />
Fundamental Optics<br />
100<br />
typical reflectance curves<br />
100<br />
typical reflectance curves<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
PERCENT REFLECTANCE<br />
80<br />
60<br />
40<br />
20<br />
0<br />
normal incidence<br />
45° incidence<br />
250 300 350<br />
400<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.46 MAXBRIte /007 coating<br />
$ Excellent performance for excimer and YAG third- and<br />
fourth-harmonic lines, as well as broadband ultraviolet<br />
sources<br />
$ Superior reflectance from 0 to 45 degrees incidence for<br />
s- and p-polarizations<br />
$ R avg > 98% from 245 to 390 nm<br />
$ Damage threshold: see /001 (similar specifications)<br />
OPTICS GLASS CLEANING<br />
PERCENT REFLECTANCE<br />
99<br />
98<br />
97<br />
96<br />
normal incidence<br />
45° incidence<br />
400 500 600 700<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.47 MAXBRIte /009 coating<br />
$ Wavelength range extended even farther than /001<br />
MAXBRIte <br />
$ Outstanding performance from 0 to 45 degrees incidence<br />
$ R avg > 98% from 420 to 700 nm<br />
$ Damage threshold: 0.4 J/cm 2 810%,<br />
10-nsec pulse (35 MW/cm 2 ) at 532 nm on silica substrate<br />
Melles Griot removes all contamination from our substrates prior to coating with a high-volume, five-stage Interlab<br />
semi-aqueous glass cleaner.<br />
5.34 1 Visit Us OnLine! www.mellesgriot.com
Laser-Line MAX-R <strong>Coatings</strong><br />
These multilayer coatings achieve the highest possible reflectances<br />
at specific laser wavelengths and at particular angles of incidence. At<br />
these wavelengths and angles, laser-line MAX-R coatings outperform<br />
MAXBRIte . We offer coatings for angles of incidence at<br />
0 degrees (see figure 5.48) and 45 degrees (see figure 5.49). Because<br />
the layer thicknesses differ for these two angles, the coatings cannot<br />
be used interchangeably. Laser-line MAX-R coatings are intended<br />
for external beam-manipulation applications. Specified coating<br />
reflectances apply to p-polarization (except at normal incidence,<br />
where the p- and s-polarization states are indistinguishable).<br />
Reflectances for s-polarization generally exceed those for p-<br />
polarization. Other coatings can be supplied at any center wavelength<br />
from 193 nm to 1.6 mm.<br />
APPLICATION NOTE<br />
LASER-INDUCED DAMAGE<br />
The following laser damage threshold statistics do<br />
not constitute a performance guarantee but should<br />
be representative for MAX-R coatings.<br />
/205 2.4 J/cm 2 810%,<br />
10-nsec pulse (195 MW/cm 2 ) at 355 nm<br />
/255 13.3 J/cm 2 810%,<br />
12-nsec pulse (920 MW/cm 2 ) at 355 nm<br />
/225 12.6 J/cm 2 810%,<br />
10-nsec pulse (1008 MW/cm 2 ) at 532 nm<br />
/275 13.6 J/cm 2 810%,<br />
10-nsec pulse (1088 MW/cm 2 ) at 532 nm<br />
/241 2.4 J/cm 2 810%,<br />
20-nsec pulse (700 MW/cm 2 ) at 1064 nm<br />
/291 17.7 J/cm 2 810%,<br />
20-nsec pulse (800 MW/cm 2 ) at 1064 nm<br />
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
Figure 5.48<br />
PERCENT REFLECTANCE<br />
100<br />
Figure 5.49<br />
normal incidence<br />
typical reflectance curve<br />
0.8 0.9 1.0 1.1 1.2<br />
RELATIVE WAVELENGTH, g = l 0 /l<br />
80<br />
60<br />
40<br />
20<br />
MAX-R coating, normal incidence<br />
45° incidence<br />
s-polarization<br />
p-polarization<br />
typical reflectance curves<br />
0.8 0.9 1.0 1.1 1.2<br />
RELATIVE WAVELENGTH, g = l 0 /l<br />
MAX-R coating, 45-degree incidence<br />
$ Highest possible reflectance achieved at specific laser<br />
wavelengths and angles of incidence<br />
$ Standard MAX-R coatings available for popular<br />
laser wavelengths at both 0 degrees and 45 degrees<br />
$ Custom designs easily produced<br />
$ <strong>Coatings</strong> available from 193 nm to 1550 nm<br />
PLEASE NOTE FOR THE ABOVE CURVES:<br />
For g = l 0 /l, l 0 is the design wavelength and l is<br />
arbitrary. For example, l = l 0 /g = 1064 nm/0.9 =<br />
1182 nm, if one looks at g = 0.9 for the /291 coating.<br />
For values of g below 0.8 and above 1.2, the<br />
reflectance curves oscillate sinusoidally, in a manner<br />
that varies from coating to coating and run to run.<br />
Fundamental Optics Gaussian Beam Optics <strong>Optical</strong> Specifications Material Properties <strong>Optical</strong> <strong>Coatings</strong><br />
Visit Us Online! www.mellesgriot.com 1 5.35
Fundamental Optics<br />
Laser-Line MAX-R <strong>Coatings</strong>, Normal Incidence<br />
Laser-Line MAX-R <strong>Coatings</strong>, 45-Degree Incidence<br />
Minimum<br />
Reflectance R p (%)<br />
Minimum<br />
Reflectance R p (%)<br />
Wavelength<br />
(nm)<br />
Laser Type<br />
0º<br />
Incidence<br />
0º±15°<br />
Incidence<br />
COATING<br />
SUFFIX<br />
Wavelength<br />
(nm)<br />
Laser Type<br />
45º<br />
Incidence<br />
45º±10°<br />
Incidence<br />
COATING<br />
SUFFIX<br />
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
193<br />
248<br />
266<br />
308<br />
351<br />
364<br />
442<br />
458<br />
466<br />
473<br />
476<br />
488<br />
496<br />
502<br />
514<br />
532<br />
543<br />
633<br />
670<br />
694<br />
780<br />
830<br />
850<br />
904<br />
1064<br />
1300<br />
1523<br />
1550<br />
ArF<br />
KrF<br />
Nd:YAG 4th harm.<br />
XeCl<br />
Ar ion<br />
Ar ion<br />
HeCd<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Nd:YAG 2nd harm.<br />
HeNe<br />
HeNe<br />
GaAlAs<br />
Ruby<br />
GaAlAs<br />
GaAlAs<br />
GaAlAs<br />
GaAs<br />
Nd:YAG<br />
InGaAsP<br />
HeNe<br />
InGaAsP<br />
97.0<br />
98.0<br />
98.0<br />
99.0<br />
99.0<br />
99.0<br />
99.3<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.2<br />
99.2<br />
99.2<br />
99.2<br />
94.0<br />
95.0<br />
95.0<br />
96.0<br />
96.0<br />
96.0<br />
99.0<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
Note: To order, append coating suffix to product number.<br />
/201<br />
/202<br />
/203<br />
/204<br />
/205<br />
/207<br />
/209<br />
/211<br />
/213<br />
/215<br />
/217<br />
/219<br />
/221<br />
/222<br />
/223<br />
/225<br />
/226<br />
/229<br />
/228<br />
/231<br />
/233<br />
/237<br />
/238<br />
/239<br />
/241<br />
/245<br />
/247<br />
/247<br />
193<br />
248<br />
266<br />
308<br />
351<br />
364<br />
442<br />
458<br />
466<br />
473<br />
476<br />
488<br />
496<br />
502<br />
514<br />
532<br />
543<br />
633<br />
670<br />
694<br />
780<br />
830<br />
850<br />
904<br />
1064<br />
1300<br />
1523<br />
1550<br />
ArF<br />
KrF<br />
Nd:YAG 4th harm.<br />
XeCl<br />
Ar ion<br />
Ar ion<br />
HeCd<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Ar ion<br />
Nd: YAG 2nd harm.<br />
HeNe<br />
HeNe<br />
GaAlAs<br />
Ruby<br />
GaAlAs<br />
GaAlAs<br />
GaAlAs<br />
GaAs<br />
Nd:YAG<br />
InGaAsP<br />
HeNe<br />
InGaAsP<br />
97.0<br />
98.0<br />
98.0<br />
98.0<br />
98.0<br />
98.0<br />
99.0<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.3<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.5<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
99.0<br />
94.0<br />
95.0<br />
95.0<br />
95.0<br />
96.0<br />
96.0<br />
98.0<br />
98.0<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.5<br />
98.0<br />
98.5<br />
98.5<br />
98.5<br />
Note: To order, append coating suffix to product number.<br />
/251<br />
/252<br />
/253<br />
/254<br />
/255<br />
/257<br />
/259<br />
/261<br />
/263<br />
/265<br />
/267<br />
/269<br />
/271<br />
/272<br />
/273<br />
/275<br />
/276<br />
/279<br />
/278<br />
/281<br />
/283<br />
/287<br />
/288<br />
/289<br />
/291<br />
/295<br />
/297<br />
/297<br />
SPHERICAL AND CYLINDRICAL MIRRORS<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Virtually any of our mirror coatings can be applied to standard optical components. Common examples include using<br />
plano-concave lenses to make spherical concave mirrors or coating plano-convex lenses to make secondary mirrors for<br />
Cassegrain beam expanders or telescopes. The radius of curvature of any of our simple lenses can be calculated using<br />
the data given in the product tables together with the formulas presented at the end of Chapter 1, Fundamental<br />
Optics.<br />
5.36 1 Visit Us OnLine! www.mellesgriot.com
Ultrafast Coating<br />
Melles Griot has developed a new coating which is designed for<br />
ultrafast laser systems in the near-infrared. The ultrafast coating<br />
(/091, shown in figure 5.50), an all-dielectric coating, centered at<br />
800 nm, minimizes pulse broadening for ultrafast applications. The<br />
coating also offers exceptionally high reflectance for both<br />
s- and p-polarizations in the range from 750 nm to 870 nm.<br />
The ultrafast coating is ideal for high-power Ti:sapphire laser<br />
applications. This coating is superior to protected and enhanced<br />
metallic coatings because of its ability to handle higher powers. For<br />
lower power light sources, the protected silver /038 coating is strongly<br />
recommended for ultrafast applications because of its low pulsebroadening<br />
effect.<br />
Melles Griot offers the /091 coating standard on both a 12.5-mmand<br />
a 25.0-mm-diameter ultraviolet synthetic fused-silica substrate.<br />
PERCENT REFLECTANCE<br />
100<br />
80<br />
60<br />
40<br />
20<br />
Ultrafast Coating (/091)<br />
Wavelength<br />
Range<br />
(nm)<br />
p-plane<br />
s-plane<br />
700 800 900<br />
1000<br />
WAVELENGTH IN NANOMETERS<br />
Figure 5.50 Ultrafast coating /091<br />
Minimum<br />
Reflectance<br />
R p (%)<br />
Angle of<br />
Incidence<br />
(deg)<br />
Pulse<br />
Broadening<br />
(%)<br />
typical reflectance curves<br />
$ R p > 99% from 770 mm to 830 nm, Rs >99% from 750 nm<br />
to 870 nm<br />
$ High laser-damage threshold<br />
$ Low pulse broadening<br />
COATING<br />
SUFFIX<br />
770–830 99.0 45
Material Properties <strong>Optical</strong> Specifications Gaussian Beam Optics<br />
<strong>Optical</strong> <strong>Coatings</strong><br />
Fundamental Optics<br />
SALT-FOG AND HUMIDITY<br />
TESTING FOR OPTICAL COATINGS<br />
Melles Griot has tested several standard optical<br />
coatings to ensure spectral performance under<br />
non-laboratory environmental conditions. The<br />
following list identifies optical coatings that passed<br />
environmental testing for salt- fog and humidity.<br />
The humidity testing was done per MIL-C-14806A<br />
paragraph 4.4.6. The salt-fog test was done per<br />
MIL-C-14806 paragraph 4.4.8 and MIL-STD-810C<br />
reference 1.3, method 509.1, procedure I.<br />
Humidity testing was done at 120ºF and the relative<br />
humidity was at 95-100% for 24 hours. Salt-fog<br />
testing was done at a temperature of 95ºF, pH<br />
solution of 6.5, collection rate of 1.18 ml/hr, and<br />
specific gravity of 1.034.<br />
<strong>Optical</strong> <strong>Coatings</strong> Passing Environmental Test<br />
for Salt-Fog and Humidity<br />
Antireflection<br />
Single layer MgF 2 /066<br />
Single layer MgF 2 /067<br />
HEBBAR /072<br />
HEBBAR /073<br />
HEBBAR /074<br />
HEBBAR /075<br />
HEBBAR /076<br />
HEBBAR /077<br />
HEBBAR /078<br />
HEBBAR /079<br />
HEBBAR /083<br />
HEBBAR /084<br />
V-coating at 248 nm<br />
V-coating at 532 nm<br />
V-coating at 904 nm<br />
V-coating at 1300 nm<br />
V-coating at 1523–1550 nm<br />
Reflective<br />
MAXBRite /001<br />
MAXBRite /003<br />
MAXBRite /009<br />
MAX-R at 351 nm<br />
MAX-R at 633 nm<br />
MAX-R at 1064 nm<br />
COATING CAPABILITIES<br />
Melles Griot uses a state-of-the-art, Eddy Company,<br />
SYS/48B ion-assisted coating chamber for our highprecision<br />
reflective and antireflection coatings.<br />
This fully automatic system produces multilayer<br />
dielectric coatings with excellent thin film quality,<br />
high damage thresholds, and low loss. In addition,<br />
the automatic coating process allows us to produce<br />
coatings with higher precision, better uniformity,<br />
and greater batch-to-batch repeatability. This benefits<br />
not only our research customers but also our OEM<br />
customers. To meet your specific needs Melles Griot<br />
can produce custom coatings in high volume or in<br />
prototype quantities. Contact your local Melles Griot<br />
sales office for more information.<br />
Our coating chamber is located in a Class-10,000<br />
clean room to ensure coatings of the highest quality.<br />
If required, we can also inspect coated optics in a<br />
clean room and then package them in special cleanroom<br />
packaging so that the parts can be shipped<br />
direct to and opened in a clean room environment.<br />
Filter/Beamsplitter<br />
Hot Mirror<br />
Cold Mirror<br />
UV plate beamsplitter<br />
03 BTF at 550 nm<br />
03 BTF at 850 nm<br />
03 BDS 001 beamsplitter State-of-the art Eddy SYS/48B coating chamber<br />
5.38 1 Visit Us OnLine! www.mellesgriot.com