Injection Molded Polymer Optics In The 21st-Century
`
`William S. Beich, Director of New Business Development, G-S PLASTIC OPTICS.
`
`1. Introduction
`Precision polymer optics, manufactured by injection molding techniques, has been a key enabling technology for
`several decades now. The technology, which can be thought of as a subset of the wider field of precision optics
`manufacturing, was pioneered in the United States by companies such as Eastman Kodak, US Precision Lens, and
`Polaroid. In addition to suppliers in the U.S. there are several companies worldwide that design and manufacture
`precision polymer optics, for example Philips High Tech Plastics in Europe and Fujinon in Japan.
`
`Designers who are considering using polymer optics need a fundamental understanding of exactly how the optics are
`created. This paper will survey the technology and processes that are employed in the successful implementation of a
`polymer optic solution from a manufacturer’s perspective. Special emphasis will be paid to the unique relationship
`between the molds and the optics that they produce. We will discuss the key elements of production: molding resins,
`molds and molding equipment, and metrology. Finally we will offer a case study to illustrate just how the optics
`designer carries a design concept through to production. The underlying theme throughout the discussion of polymer
`optics is the need for the design team to work closely with an experienced polymer optics manufacturer with a solid
`track record of success in molded optics. As will be seen shortly, the complex interaction between thermoplastics,
`molds, and molding machines dictates the need for working closely with a supplier who has the critical knowledge
`needed to manage all aspects of the program.
`
`2. Applications
`Starting in the middle of the last century optics were molded
`for simple condenser lenses, toy objectives, ophthalmic
`applications, gauge windows and other low quality glass
`optic replacements1. At the dawn of the 21st Century
`polymer optics can be found in many more sophisticated
`applications. Examples can be offered to illustrate the
`widespread versatility of polymer optics and optical systems.
`The spectrum would range across medical, military and
`commercial applications.
`
`2.1. Medical
`Medical instruments such as laparoscopes and arthroscopes
`can be built using polymer singlets and doublets, sometimes
`in conjunction with traditional glass elements to correct
`optical aberrations. Polymer optics can be used for non-
`imaging applications in medical devices as well.
`
`2.2. Military
`Imaging systems such as those found in night-vision devices
`Fig.1: ITT Night Vision
`Goggles. The PVS-7 is
`are a good example of how the properties of polymers can
` designed and manufactured by
`combine to address several key performance issues (see
`ITT Industries Inc.
`Figure 1). This example illustrates how a lightweight
`wearable night vision device containing as few optical elements as possible can be designed and manufactured because
`of the sophistication of the molded optics. Replacing a glass element with a plastic element can reduce weight in the
`system. A plastic element is approximately a factor of 2 to a factor of 5 lighter in weight than the glass element being
`replaced. Moreover, since polymer optics can be readily designed and manufactured with aspheric surfaces (resulting in
`a possible reduction in the total element count) the use of one or more polymer elements in the system is well advised.
`
`Tribute to Warren Smith: A Legacy in Lens Design and Optical Engineering, edited by Robert E. Fischer,
`Proc. of SPIE Vol. 5865 (SPIE, Bellingham, WA, 2005) · 0277-786X/05/$15 · doi: 10.1117/12.626616
`
`Proc. of SPIE 58650J-1
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 1 of 12
`
`

`

`2.3. Commercial
`The use of polymer optics is found in a wide range of consumer
`applications
`such as digital cameras, PC peripherals,
`videoconferencing cameras, and mobile imaging. Laser based
`bar code scanners of various descriptions are good candidates for
`polymer optics (see Figure 2). More recently, lightweight image
`based scanners have been manufactured to read 2-D barcode
`images using a platform that contains multiple optics and a
`mechanical mount all in one shot (see Figure 3). Biometric
`security systems, smoke detector optics, automated flush valve
`systems, and laboratory equipment have all benefited from
`having precision polymer optics. Polymer optics are useful for
`certain telecom and datacom products and are commonly used to
`replicate micro structured surfaces such as Fresnel lenses,
`refractive-diffractive surfaces, and gratings.
`
` Fig.2: Microscan Barcode scanner
` and polygon mirror. The MS-820
`Scanner is designed and manufactured by
`Microscan Systems Inc., Renton, WA.
`3. Elements of Production
`We will now turn our attention to the various elements that are required to manufacture precision polymer optics. They
`can be distilled into three major factors: thermoplastic resins, molding machines and molds, and metrology. Noticeably
`absent from this list is the optical design. A robust optical design is of course required but is beyond the scope of this
`discussion. What we will discuss is the importance of having the optic thoroughly designed for manufacturability given
`that the final product is a polymer optic. When considering polymer optics the need for a manufacturable design is
`vital. Manufacturability of design is important when considering glass designs to be sure. But when polymer designs
`are being considered, an optical design that works well in Zemax or Oslo will fail the designer if the manufacturing
`process itself is ignored. There is simply no way to gloss over this fact.
`
`From a manufacturer’s perspective this is often seen when glass
`designs are converted into the plastic analog (for example, that BK-7
`element, n=1.5151, now becomes a PMMA element with n=1.49).
`The temptation is to plug in the new index numbers, alter the radii of
`curvature accordingly, and think that one is all set for production. In
`some instances, it is possible to simply change the index (or some
`other material related value) to accommodate the new material and let
`the design software make the necessary changes. In other instances
`the form of the optic is altered to the point where it is no longer
`manufacturable, for example if the radius of curvature becomes very
`steep or if the diameter to center thickness ratio becomes too extreme.
`With guidance from the manufacturer, it may be possible to design
`the optic so that it can be produced by the injection molding process.
`In this brief example converting the index to that of the new
`thermoplastic resin is a necessary but not sufficient step in converting
`an element from glass to plastic.
`
` Fig.3: Code 2-D Scanner and
` integrated optic. The Code
`Reader 2.0 (CR2) is designed and manu-
`3.1. Polymers
`factured by Code Corporation, Draper, UT.
`In general two types of plastic materials are molded: thermoplastic resins, which can undergo repeated cycles of heating
`and cooling and thermoset resins, which become permanently insoluble after they have reached their final heating
`temperature2. This article will not discuss the thermoset resins. Thermoplastic resins are used for injection molding
`polymer optics.
`
`The most commonly used optical thermoplastics are: Acrylic (Altuglas, Cyro), Polystyrene (DOW Corning) and
`Polycarbonate (GE Plastics, Bayer). More recently there are the Cyclic Olefins such as Topas (Ticona) and Zeonex
`(Zeon Chemicals) and specialty resins such as polyetherimide, with the trade name of Ultem (GE Plastics). All of these
`materials have properties and characteristics that need to be considered given the intended application. With the
`
`
`Proc. of SPIE 58650J-2
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 2 of 12
`
`

`

`exception of Ultem most are suitable for broadband visible applications, with transmission falling off steeply below
`350nm and above 1600nm. Acrylic and Zeonex have excellent external transmission properties in the visible (around
`92%). Polycarbonate and Polystyrene have slightly more haze (with external transmissions of about 87% to 88% in the
`same region). The coefficient of thermal expansion of the polymers is approximately an order of magnitude higher than
`that of glass (7 or 8 X 10-5/o C). Most polymers have a fairly low maximum sustained operating temperature (MOT)
`(the temperature at which the element can survive for extended periods of time without deformation). This ranges from
`about 80o C in Polystyrene to about 185o C in Ultem. Ultem is a special case material, in optical molding. It is not
`suitable for broadband visible applications and represents the upper limit in terms of temperature survivability; most
`Polycarbonates and Olefins have a MOT of about 115o to 125o C. With the exception of the Cyclic Olefins, most
`polymers will absorb water and change dimensionally. Acrylic for example will absorb 0.3% H2O (immersion at 23o C
`for 24 hours) while the COP Zeonex E-48R resin will absorb <0.01%. Moreover the change in index of refraction with
`temperature is fairly large (about 20 times that of glass) and negative3. Maintaining focus over a wide temperature
`range is a significant problem for plastic optics3. Another challenge for designers using the optical thermoplastics is
`that the range of index of refraction is fairly narrow. Acrylic is about 1.497, Polycarbonate about 1.599, Polystyrene
`about 1.604, the Cyclic Olefins around 1.53 and finally Ultem at 1.689. Table 1 below summarizes these and other
`important properties.
`
`Table 1: Important Properties of Selected Optical Grade Polymers
`Values may vary significantly depending upon manufacturer4.
`
`
`Optical injection molders have no control over the properties of the thermoplastic resins. Questions about the
`mechanical properties of the thermoplastic resins, the precision of the index of refraction, and questions about the
`consistency of material from lot to lot are best referred to the resin suppliers.
`
`
`Proc. of SPIE 58650J-3
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 3 of 12
`
`

`

`3.2. Injection Molding Machines
`The injection molding machine consists of a clamp unit at one end, an injection or plasticizing unit at the other end and
`a mold area in the middle. The control station is usually located near the injection unit. Figure 4 illustrates the concept
`of a typical injection-molding machine. There are obviously many different types and configurations of molding
`machines. For the purposes of our discussion we will consider a machine in horizontal configuration, using an
`hydraulic clamping unit and single-stage plasticator.
`
`Molding machines can
`range in size up to 10,000
`tons of clamp force. As
`clamp tonnage is dictated
`to a
`large measure by
`projected area, for
`the
`optical molding world,
`machines tend to run from
`20 to 300 tons of clamp
`force.
` Some micro
`molding applications call
`for machines with lower
`tonnage.
`
`The mold is mounted on
`the platens which are part
`of the clamp unit. One of
`the platens
`is fixed in
`place,
`the
`other
`is
`moveable.
` There are
`typically four tie bars supporting this area of the press. During the molding cycle the clamp unit closes the moveable
`platen against the fixed platen. The mold itself is split into two halves (fixed and moveable). In effect the mold is
`separated across the parting line to allow for removal of parts. The clamping unit also exerts enough force on the mold
`halves to ensure that they are not forced apart during the injection phase. There are various types of clamping schemes:
`hydraulic, toggle, and hydromechanical.
`
`The injection unit consists of a hopper located over the reciprocating screw and barrel. The reciprocating screw is
`located within the barrel. Heater bands surround the barrel. The thermoplastic resin is melted (a process known as
`plasticizing) using a combination of conductive heat from the heater bands and frictional heat from the rotating screw.
`Please note the unusual construction of the screw. See Figure 5. The screw begins to turn to collect material for the
`next shot. As the screw turns, resin from
`the hopper flows into the chamber and is
`collected in the feed section of the screw.
`The material
`is
`trapped between
`the
`chamber wall and the screw. The chamber
`wall is the bearing surface where shear is
`imparted to the melt as it is fed forward on
`the screw into the transition region of the
`screw. As the melt moves forward it
`accumulates at the tip the screw. As a result
`the screw is driven backwards. At a preset
`point the mold closes and the screw stops
`turning. The screw then moves forward,
`acting like a ram, injecting all the melt into
`the mold in a single stage.
`
`
`Fig.4: Injection Molding Machine
`
`Fig.5: The Screw
`
`
`Proc. of SPIE 58650J-4
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 4 of 12
`
`

`

`The molding machine is designed to allow for complete control of the shot size, injection speed, injection pressure,
`backpressure, cushion and a matrix of other variables that ultimately determine the outcome of the molded element.
`Once the resin has been held in the mold for a sufficient time and has cooled, the screw starts rotating again to prepare
`for the next shot. The mold opens and optics are removed.
`
`Other equipment is frequently added to the molding press as required to facilitate operations. For example the hopper
`on most machines can only hold a limited amount of material. Automated hopper feeds can be added to ensure a
`continuous feed of dried resin. Desiccators are used to remove moisture from the resin prior to molding. It is important
`to remove the shot from the press on cycle without fail. Robotic sprue pickers can be added to handle this task. It is
`possible to add other end of arm tooling to the robots to automate degating and placement of elements into appropriate
`packaging. Degating is the process whereby the elements themselves are removed from the runner system and placed in
`trays or holders. All of this tooling is driven by the program requirements.
`
`3.3. Molds
`The mold used to manufacture polymer optics can be thought of as a sophisticated three dimensional puzzle that has two
`main features: (1) the cavity details along with the core pins (also known as optical inserts or nubbins), and (2) the
`frame (sometimes called the base) that houses the cavities and inserts. Figure 6a and 6b illustrate the basic concept of
`the mold. The complexity of the mold is primarily driven by the complexity of the element being molded. One of the
`key advantages of using polymer optics is the ability to combine optical and mechanical features into one platform.
`Depending upon the nature of the mechanical features being considered the mold itself will take on additional
`complexity. The following table is a brief list of some of the key features generally found in molds along with a brief
`description of their intended function 2.
`
`
`
`
`
`
`
`
`
`
`
`
`Fig.6a: The Mold.
`
`
`Proc. of SPIE 58650J-5
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 5 of 12
`
`

`

`Fig.6b: Actual Mold
`Parting Line View
`
`
`
`ITEM NO.
`1
`2
`3
`4
`5
`6
`7
`
`
`
`MOLD FEATURE
`Sprue Bushing
`Top (A-side) plate
`Guide Pins
`Bottom (B-side) plate
`Ejector Mechanism
`Ejector Housing
`Runner System
`
`8
`9
`
`10
`
`Vents
`Gates
`
`Optical inserts (nubbins)
`
`FUNCTION
`Provides means of entry into the mold interior.
`Portion of the mold mounted on the stationary side of the press
`Maintains proper alignment of the two halves of the mold.
`Side opposite the “A” side, sits on the moveable platen.
`System used to eject rigid molded elements from the cavities.
`Houses the ejector system.
`System of channels in the mold face used to convey molten plastic from
`the sprue to the cavities.
`Structure that allows trapped gas to escape.
`Region of the mold that controls the flow of molten material into the
`cavities.
`Deterministically ground and polished steel or steel/nickel pins against
`which the optical surface forms.
`
`
`Table 2: Mold Features
`
`The mold is mounted into the molding press. One side of the mold is locked into the stationary side of the press; the
`other side of the mold is mounted on to the moveable platen. During molding the two halves of the mold are closed and
`clamped under high pressure. The molten resin is injected into the mold from the molding press injection unit (also
`called the plasticator) located on the stationary side of the press. The melt flows through the passages within the mold
`(that is the sprue and the runner) and enters the cavity through the gate. The cavities fill with resin and take on the
`shape of the cavity detail and the form of the core pin. As the resin is held within the mold it is cooled to an appropriate
`temperature at which time the mold opens (the moveable platen on the molding press moves away from the fixed
`platen) and the ejector system is engaged to remove the elements from the mold. The elements are usually part of the
`runner system and this is set aside to allow further cooling and subsequent finishing.
`
`When molding optics, much thought and effort is put into creating the cavity impression and the optical inserts in such a
`fashion that the final molded product will be at the proper dimensions at room temperature. All of the thermoplastic
`resins exhibit a measure of shrinkage as they cool in the cavity. A typical shrinkage would be 0.005 inch per inch of
`linear dimension. This shrinkage factor must be taken into consideration when laying out the cavity details. If not, the
`resultant elements will be too small. Note: if the element is very small, shrinkage factors become less of an issue.
`
`
`
`Proc. of SPIE 58650J-6
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 6 of 12
`
`

`

`Whenever the final part dimensions are very critical, an iterative process may be used to achieve the final dimensions.
`Initial shrinkage calculations are made and the mold will be built tool-safe. That is, the mold will be built slightly
`smaller than is required. After an initial molding trial has been completed, and the molded elements characterized, the
`actual shrinkage is better understood. Steel can then be removed from the mold cavity to increase the size of the
`element in a deterministic fashion. This process applies as well to the creation of the optical insert. Hence it may be
`necessary to rework the optical insert to correct for form error after an initial molding trial is undertaken and the
`molding process is understood.
`
`The inserts can be made in several ways. One way is to grind and polish fine-grained steel in a fashion similar to the
`techniques employed in fabricating conventional glass optics. Another method used to make the insert is to fabricate a
`steel blank, plate it with an electroless nickel alloy and subsequently diamond machine the surface. Today single point
`diamond machining technology is capable of extremely accurate surfaces both in terms of surface error and RMS
`surface roughness. For example on a 1” diameter optical insert (steel base with nickel plating) a wavefront error on the
`order of λ/2 with a surface roughness of between 70Å to 50Å RMS is achievable. This can be achieved without post-
`polishing the surfaces.
`
`3.4. Metrology
`Having good metrology is an absolute requirement when molding optics. It is really no different than what is required
`for traditional glass fabrication. Without adequate metrology there is no reasonable way to assess performance.
`Molded optics can be tested on an interferometer. If the surface is aspheric, null correctors can be designed and
`employed. Another useful device is a contact profilometer. Many contact profilometers were originally designed for
`surface roughness measurements. However with the addition of solver programs it is possible to trace aspheric surfaces
`and measure form errors. Because of the common practice of integrating mechanical and optical features onto one
`platform, appropriate metrology is required to ensure these features are to print. Micrometers, calipers, optical
`comparators, and microscopes are appropriate for this. A coordinate measurement machine is also useful for this
`application. This is especially true if the molded optic has optics in multiple planes relative to a datum, such as is found
`in a rotating polygon.
`
`Since injection molding is a volume related process, the use of some kind of functional test equipment is often required
`and can be tailored for the program needs. Functional testers allow for a go/no-go assessment of the process in a timely
`fashion. The downside to using this type of metrology is that when problems do arise the functional diagnostic will not
`necessarily allow the QC technician to uncover the root cause of the problem. The operator may be aware of a problem
`so lot containment can take place, but often other QC equipment will be required to push for the root cause problem.
`Additional equipment that can be used would include a lens bench for measuring flange focus, an MTF diagnostic, and
`resolution targets.
`
`4. The Decision to Use Glass or Plastic
`The question is often asked, “Why use plastic? Why not just stay with glass?” The answer is that some applications are
`simply better served by using a plastic element instead of a glass element. As noted earlier, the use of plastic optics
`today is found in a wide array of military, medical and commercial applications. There are some general guidelines that
`can be followed that will help determine when it is appropriate to consider a polymer optic solution.
`
`4.1. How do the thermoplastic resins affect the decision
`The factors that would argue against using polymer optics in an application are almost always centered on the
`limitations inherent in the thermoplastic resins themselves. For example: (1) spectral transmission (most polymers are
`suitable for the visible portion of the spectrum), (2) continuous service temperature (lower than glass, overall less than
`120o C), (3) dn/dt (close to a factor of 20 higher than glass and negative) 3, (3) coefficient of linear expansion (about an
`order of magnitude higher than glass) 3, (5) surface hardness (softer than glass), and (6) index of refraction (the index
`map is limited relative to glass). These properties do not preclude their use in optical systems, however the designer
`must be well aware of their performance limits when considering their use. It should be noted that not all of the
`properties of the thermoplastic resins are limiting. One prominent characteristic of thermoplastic resins that would be a
`desirable indicator for use in a system is the specific gravity of the material. Overall, polymers are lighter in weight
`than their glass counterparts, a positive factor when designing a weight-sensitive system.
`
`
`Proc. of SPIE 58650J-7
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 7 of 12
`
`

`

`4.2. How does the manufacturing process affect the decision
`A key factor for considering polymer optics is the method of manufacturing itself, namely injection molding. Injection
`molding is a highly efficient method of reproducing optics with complex surface geometries. Moreover, such optics can
`employ integrated mounting features onto one platform and can be molded in varying volume requirements with a very
`high degree of part-to-part repeatability. The reason for this is the fact that the molds are built to a higher degree of
`precision than is required in the part.
`
`When considering molds and the molding process the key thing to understand is that the tooling, while expensive, is
`usually done once and if done properly will last a very long time. It is not uncommon to get over 1,000,000 cycles on a
`fully hardened mold. Effort is put into building the best mold (base, cavity and optical inserts) to the highest tolerances
`with replication taking place off of the mold. So instead of each element being treated as a one-off component,
`replication is taking place within a highly precise master tool. This is where great economies of scale can be realized
`because it is possible to build multiple cavities into one mold base. If (1) the application involves a fairly high volume
`of components, or (2) if the optic has a combination of optical and mechanical features that are to be integrated onto one
`platform, or (3) if the optical surface is aspheric or if the optic is very complex, such that building a mold and running
`low production volumes is still more cost effective than competing options, then polymer optics would be a good
`consideration because of the method of manufacturing, injection molding.
`
`4.3. How the shape of the element affects the decision
`While the injection molding process is highly efficient, and the molds themselves are built to very high tolerances, the
`molding process is not without its own set of challenges. These challenges are often related to the shape of the optic
`and the fact that the optic may not be optimized for manufacturing. As noted earlier problems tend to arise when the
`optical design is done without considering the manufacturing process.
`
`Some examples of difficult to mold optical
`forms would include: (1) a biconvex lens that
`has a very thick center and a very thin edge, (2) a
`bi-concave lens with a very thin center and a
`very
`thick edge, (3) a plano-convex
`lens,
`especially where the convex surface is quite
`strong, (4) an element with very stringent surface
`figure error or irregularity requirement, (5) an
`element whose clear aperture is equal to the
`geometric size of the optic, and (6) an optic with
`mounting
`feature
`requirements
`that
`are
`incompatible with molding a good optical
`surface (for example a mounting boss located
`directly behind a plano facet). Figure 7 shows
`some shapes that are very difficult to mold.
`
`Fig.7: Challenging shapes to mold
`
`4.4. Limits of Fabrication
`Manufacturers are frequently asked the question, “How good can you make it?” The answer of course is, “It depends.”
`It depends on many factors. Several of these have been hinted at in the discussion above on when it is appropriate to
`consider using a polymer optic. When discussing the issue of the limits of fabrication closer attention needs to be paid
`to the form of the optic as the injection molding process will affect the final product. As noted above, elements that
`have extremes in wall thickness are more difficult to mold than elements of nearly uniform wall thickness. In terms of
`surface form error, square or rectangular shapes are more difficult to mold than round shapes. Figure 7 summarizes the
`discussion of optical forms that are challenging to mold.
`
`One thing is clear. As the design of the optic progresses to the stage where the question of limits of fabrication are
`being asked, it is important for the designer to contract with an experienced optical molder for guidance through the
`fabrication process. As the design progresses, the molder will provide critical guidance to the designer on issues of
`manufacturability. As noted above this is often an iterative process as manufacturing features are traded off against the
`
`
`Proc. of SPIE 58650J-8
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 8 of 12
`
`

`

`requirements of system performance. When discussing issues of manufacturing tolerances ultimately the question
`becomes one of what constitutes a manufacturable design. Again, it is very difficult to generalize beyond some basic
`rules of thumb; each design presents its own challenges and must be approached on a case-by-case basis. Other factors
`enter into the discussion at this point. For example, the choice of materials (when more than one is available), the best
`method for tool construction (the number of cavities in the mold, the type of mold base and the proper cooling
`configuration), and the way the part is gated (this is critical when stress birefringence is an issue).
`
`Also, from a manufacturing perspective, this is the point in time when the optical molder needs to discuss such issues as
`degating and finishing the element, what quality metrics will be used to insure good consistent production and what
`kind of packaging needs will the program have. Table 3 provides a rule of thumb guide describing the state-of the art
`limits of fabrication. Each optic needs to be evaluated on an individual basis.
`
`
`ATTRIBUTE
`
`Radius of Curvature
`EFL
`Center thickness
`Diameter
`Wedge (TIR) in element
`S1 to S2 Displacement (across the mold parting line)
`Surface Figure Error
`Surface Irregularity
`Scratch-Dig Specification
`Surface Roughness Specification (RMS)
`Diameter to Thickness Ratio
`Center Thickness to Edge Thickness Ratio
`Part to Part Repeatability (one cavity)
`
`Table 3: Limits of Fabrication4
`
`
`TOLERANCES
`(ROTATIONALLY SYMMETRICAL ELEMENTS
`LESS THAN 75MM IN DIAMETER)
`
`± 0.5%
`± 1.0%
`± 0.020mm
`± 0.020mm
`< 0.010mm
`< 0.020mm
`≤ 2 fringes per inch (2 fringes = 1 λ)
`≤ 1 fringe per inch (2 fringes = 1 λ)
`40-20
`< 50 Å
`< 4:1
`< 3:1
`< 0.50%
`
`4.5. Polymer Optics and Coatings
`Polymer optics can be coated using physical vapor deposition techniques, similar to glass. It is possible to apply
`dielectric coatings such as broadband AR, V-coatings, and special band pass coatings. Moreover, reflective coatings
`can be applied to polymer substrates. Coatings such as enhanced or protected gold, aluminum and silver are routinely
`applied. It should be noted that these coatings are applied at lower temperatures and are generally not as robust as
`coatings found on similar glass elements.
`
`5. A Case Study: How to Work with an Optical Molder
`The following case study will illustrate how the process of implementing a polymer optics solution works.
`
`Irwin Industrial Tools first approached us about their requirement in January 2004. At this point there was no design
`only a concept: Irwin wanted to create a hand held consumer product that would project two orthogonal lines when held
`in position against a wall. The product would launch nationwide through major hardware store distribution channels the
`following Christmas season. We were given information about the laser sources and the overall package constraints
`and were asked if it was possible to develop a polymer optic design to satisfy this application. We thought that it would
`be possible.
`
`Working with one of our design partners, we developed an optical design that would generate two orthogonal lines with
`a certain line characteristic over a certain length from the source. The goal for the program was to have the optical
`element molded as one piece. We were able to achieve the goal. The following process was followed to allow for
`careful implementation of the new design: (1) Develop initial optical design concepts. (2) Evaluate designs and choose
`
`
`Proc. of SPIE 58650J-9
`
`Apple v. Corephotonics
`IPR2019-00030
`Exhibit 2021 Page 9 of 12
`
`

`

`one. (3) Take one design and optimize it for optical performance and manufacturability. (4) Hold a final design review.
`(5) Prototype the design using diamond machining. (6) Evaluate optical design performance and provide feedback to
`designer if last minute changes were necessary. (7) Build a prototype one-cavity mold that would serve as a test bed to
`develop the molding process. Metrology was put in place during this phase. (8) Build a multicavity production mold
`using what was learned from the one cavity mold experience. (9) Validate each element off of the production mold,
`release to production.
`
`The design phase was fairly typical for this
`kind of project: an optical designer worked on
`the ray tracing aspects in optical design
`software such as Zemax. Three concept
`designs were presented. One was selected as
`the closest fit to the program goals and was
`carried through an optimization phase. As the
`design was nearing completion we got
`involved reviewing the mechanical and optical
`tolerances. Given the form of the element, we
`estimated what manufacturing tolerances might
`be achievable and provided detailed guidance
`to the product designer and the optical engineer
`on issues affecting manufacturability. This
`included gate location, tooling layout, choice
`of thermoplastic resin,