Leading-edge production technologies behind Canon's high-performance lenses

Spherical lenses possess an inherent flaw, in that it's impossible to converge parallel light completely onto a single point. A spherical lens has a rounded surface, which makes light passing through the edges of the lens gather at a point closer to the center of the lens compared with light passing through the center of the lens. The resulting image is slightly blurred; a problem we call "spherical aberration." To correct this spherical aberration, multiple lenses had to be combined in layers. The aspherical lens was developed as a one-piece solution for eliminating spherical aberration.

As the term suggests, the surface of the aspherical lens is "not spherical." Its curvature is slightly changed in the radial direction, and if you look at a cross-section of the lens, you'll find some depression in the lens rim. This shape enables light passing through the lens to converge onto one point, thereby producing a sharp, clear image.

The basic concept behind processing a spherical lens by grinding and polishing is to transcribe the exact shape of the tool onto the lens material. In other words, transcription is done by rubbing together the tool, which is the inverted shape of the lens you want to make, with the lens material.

However, an aspherical lens has a complexly curved surface, making it impossible to create such a lens by simply rubbing the tool and the lens material together.

In the grinding and polishing method, we use a whetstone, which is a tool incorporating diamonds. The whetstone grinds the lens material as it turns so as to carve out the complex shape of the required aspherical surface. In this case, the precision of the processing machine, which controls the position of the lens and that of the whetstone, directly affects the precision of the lens being made. Thus, an extremely precise lens requires an equally high-precision processing machine.

First and foremost, the tool must be of high precision. In the grinding process, diamonds do their job so powerfully that the lens surface comes to look milky. In order to obtain a lens with high transparency, it's vitally important to properly manage the shape of the tool and the conditions of diamonds. It's true that the larger the tool diameter, the higher the tool's efficiency. However, if you use a tool larger than the lens size, it will make no difference to the grinding and polishing method for spherical lenses, and you're left with an ordinary spherical lenses. Conversely, the smaller the tool size, the easier it is to process complex aspherical shapes. However, that means longer processing times. To achieve the ideal balance between processing time and aspherical grinding and polishing precision, we need a special tool design technology.

The next key point is how you operate the processing machine. You won't achieve the required level of precision if you leave everything to the precision of the processing machine. The most challenging part of achieving the required lens precision is to work out an ideal combination between the processing machine, tools and the software that controls the movement of the processing machine. You cannot determine the ideal combination of these three factors unless you identify the causes of errors that result in discrepancies between the lens' design and the final shape after processing. All of us--the members responsible for manufacture at the plant, the equipment development and process development teams--put in countless hours of hard work. I'm sure our success in achieving the precision of our aspherical lenses owes much to our tireless efforts.

This method uses a precision-fabricated aspherical mold which is directly pressed against glass to form a lens. After the glass and the mold for transforming the glass have been set in position, the glass undergoes heating, pressing and cooling processes before it is taken out of the mold.

This method's merit lies in its high productivity. It allows us to easily exchange different molds, making it easy to produce various types of lenses.

Glass comes in various types and each type has its own chemical composition, so you can't process the different types of glass using the same method. Shape also has a significant effect on the degree of difficulty involved in processing. Therefore, we have to design a process according to the type of glass and its shape. This is the major difficulty involved in the glass molding method.

Despite the steady progress being made in glass molding technology, there's still much that we don't understand. For example, it is difficult to quantify what exactly will happen when the glass is pressed under high temperatures, or when the glass is cooled afterward.

One day I was struggling with trial and error. It was then that my senior told me, "You have to imagine yourself as the glass." At the time, I thought they were half-joking. But soon after that, I actually tried to imagine what shape the glass would want to take when a load is applied to it. I came to realize the importance of thinking about such problems from the viewpoint of the glass. When I finally managed to solve the problem, I felt as if I had come to understand the feelings of the glass a little better, which made me happy (laughs).

There are many types of lenses. For example, some have only convex surfaces or concave surfaces on both sides, while others have a convex surface on one side and a concave surface on the other. Of course, the way a convex lens comes off the mold differs from that of a concave lens. This is why we needed to develop lens manufacturing processes according to lens shapes.

While we can anticipate to some extent such shape-related difficulties in the design stage, overcoming these obstacles in the development stage involves continuous trial and error.

Light refracts when it passes through a lens. And when encountering an obstacle, light bends around behind the other side of the obstacle. This phenomenon is known as diffraction. Using elements called "Diffractive Optics," DO lenses are designed to correct chromatic aberration by means of both refraction and diffraction of light.

Chromatic aberration in common refraction lenses occurs in a completely opposite manner to that of DO lenses. With the former, images form in the order of Blue → Green → Red, whereas with DO lenses, images form in the reverse order: Red → Green → Blue. By combining these two completely different lenses, we were able to negate the two contradictory chromatic aberrations. As a result, we are now able to create telephoto and other lenses that are significantly smaller and lighter.

Canon produces multi-layer (dual-layer and triple-layer) DO lenses and gapless dual-layer DO lenses.

Multi-layer DO lenses come in two kinds of different structures. The dual-layer type has an air layer sandwiched between two DO elements. With the triple-layer type, one of the two DO elements is coated with a different material. As product development advanced from dual-layer to triple-layer types, it became possible to use DO lenses in zoom lenses in addition to fixed-focal-length lenses.

The structure of the gapless dual-layer type is much simpler; A DO element is placed between two materials, leaving no gap. A special resin or resin material is used to make the DO element of this structure, which made it possible to diffract light rays effectively over the entire visible spectrum.

There were actually many things that proved difficult, like how to handle such a unique material, and devising a method to thin it.

The lattice structure is formed by spreading the material on the mold and then curing it using UV light. However, we encountered various difficulties in designing this process, namely figuring out how to make the lattice structure form exactly according to specifications. We knew the answers to these problems in theory, but unpredictable phenomena sometimes occurred during actual trials. Solving these issues and making the technology work was a real challenge.

But although it was a struggle, I felt it was that much more worthwhile and rewarding. Canon's technological portfolio is vast and diverse, but processing technology alone will not enable a company to create a lens like this. Nor will simply having the materials. Canon's strength lies in the fact that we have possess both, and I'm proud to be a part of that.

Natural light, with its wave-like properties, consists of an optical spectrum made up of different wavelengths, each with its own refractive index. Therefore, ordinary lenses are unable to focus light on a single point to produce a clear image. A variety of optical elements have been developed to date for lenses in an attempt to correct the chromatic aberrations that occur as a result. However, especially difficult was the development of lenses that can focus blue light, which is in the short wavelength range.

The BR optical element has the ability to target blue light alone and significantly refract it. As such, lenses incorporating this optical element can correct chromatic aberrations at an incomparably higher level than conventional lenses. So much so, it has become possible to obtain high-quality images with the least amount of blur even in astrophotography - a field where expressing each point has traditionally been extremely difficult.

We were searching for a solution to correct chromatic aberrations that the lenses available at the time designed using existing technologies couldn't. It was then that our Design Division suggested we think about a method that could significantly bend blue light. So we began by looking into whether a material capable of that existed.

Of the many materials available on the market, some were already known to be able to significantly bend blue light. However, we had to give up on them because they were difficult to apply to lenses. Then we made up our mind to develop our own original material from scratch so that we didn't have to compromise in any respect.

Since this was our first attempt at creating an organic material in-house starting with its molecular structure, we began by setting up a lab.

Then we proceeded to material simulations and actually synthesizing candidate materials to evaluate their properties. We repeated this process over and over again. With a material simulation, regardless of whether a material could be actually synthesized or whether a material could be stable as a substance, as long as you input a value it will invariably come out with a result. If you merely fixate on the desired properties, however, a simulation could lead to unthinkable molecular structures that you have no way of producing. Therefore, we gave thought to the development in various ways: whether a material of this molecular structure could be synthesized this way, whether a material of that molecular structure could be stable under various conditions, and so on. In this way, we narrowed down candidate materials by actually synthesizing them for evaluation.

Whenever I reached an impasse, I would go through the catalogues of material and reagent manufacturers page by page while trying to come up with new ideas.

While people are likely to pay attention only to the material’s optical properties like its ability to bend blue light, we actually had to design the material taking into account various other factors as well, such as its manufacturing process, safety for the human body, workability when handling the material, and long-term stability. These were integral to the “unprecedented performance” that we sought in designing the material. In this sense, I can proudly say this material embodies our unparalleled enthusiasm for and commitment to quality and performance.

The project was actually nothing less than a series of slow and steady efforts, but it was fun as well as challenging and rewarding. When I managed to find the solution to a given problem and achieve sought-after performance, I was practically shouting "eureka!"

The sense of achievement in creating something innovative that nobody had ever made before was the greatest experience I've had. Since a considerable part of this development project was personally entrusted to me, I did my best to accomplish what I had really wanted to do as a materials developer, instead of merely doing what I was told. This motivated me very much.

The field of new materials development is one where you can discover and understand things only after you have given it a try. I'm the kind of person who needs to try something and see for myself that it doesn't work before I can accept it, so I went out of my way to do things with my own two hands and confirm everything with my own two eyes.

That's true. The EF35mm f/1.4L.II USM lens of course incorporates several important improvements, but above all, the BR lens allows it to outperform other lenses in its ability to correct chromatic aberration. Eyeing its expanded application in more products, we have already begun R&D efforts for further sophistication and cost reduction of the BR lens.

To enhance the anti-reflective performance of our lenses even further, we took up the challenge of developing a cutting-edge lens coating technology using nanotechnology.

The key to preventing the reflection of light is how to control the boundary surface between the outside air and the thin film. In 2008, we began to apply SWC (Subwavelength Structure Coating) technology to our lens products for the first time. SWC is a ground-breaking technology in that it utilizes countless wedge-shaped structures of nanometer size (even smaller than the visible light wavelength: 380-780nm^*) arranged on the lens surface so that light does not react to the boundary surface in effect. This coating also exhibited stellar results for types of light whose reflections were previously difficult to control, such as light entering at an angle, significantly reducing the occurrence of ghosting and flaring.

However, the making of nano-sized wedge-like structures particular to SWC requires highly sophisticated, complex processes. Accordingly, our next target project was to develop a high-performance anti-reflective film, the production method of which would be much simpler and require no such difficult-to-make structures as SWC's. This is how the development of ASC (Air Sphere Coating) technology began.

• * 1nm (nanometer) = one billionth of a meter

In principle, wedge structures don't form a boundary at all whether light comes in from above or sideways. If there is no boundary, then the lens can demonstrate outstanding anti-reflective performance. As such, SWC delivers an exceptional performance not only for light coming in diagonally, but also for light coming in almost vertically. This coating is not specifically for light coming from a certain angle but can effectively deal with light coming from any direction.

On the other hand, ASC has a boundary surface. This means the ASC-coated film produces a superior effect on light coming from a near-vertical angle while SWC exceeds ASC in anti-reflective performance for light coming in diagonally.

We developed the layer with an ultra-low refractive index for the purpose of minimizing reflection on the boundary between the outside air and the film. In the case of films made of ordinary materials, the average refractive index for films with low refractive indices is around 1.37. Given the refractive index of air is 1, in order to minimize light reflection as close to that of air as possible, we came to the conclusion that having the film contain air inside it would be the only solution, save for the idea of using wedge structures to eliminate the boundary between outside air and the film, as is the case with SWC. Then comes the problem of how we can have the film contain a good amount of air within it. In the case of ASC, we adopted the method of having the film contain countless fine particles of air.

If the size of each air particle is too large or if air particles are distributed unevenly, air itself could become a cause of diffused reflection, degrading the transmittance of light and making the film useless as an anti-reflective film. On the other hand, too large an amount of air contained can make the film vulnerable.

So, you could say that the key point of the ASC coating technology is to control the size and quantity of air particles by giving full consideration to the materials in terms of both their composition and the manufacturing method used. As a developer involved in both materials and process development, I went to great lengths to hammer out a solution from both angles while taking a good balance between performance capabilities and how the lenses would handle into consideration.

In a sense, the development of an anti-reflective film is a seemingly less-exciting job because it concerns an invisible nanometer world; the better performance the film delivers, the less conspicuous its existence becomes. However, let's take the commercialization of the recently launched “EF11-24mm f/4L USM” lens for example - a lens with a super-wide-angle focal length. Making possible image capture that was previously impossible, to be taken now, the ground-breaking technology behind this fantastic lens is sure to have a real impact on the culture of photography. I was especially happy when I saw how satisfied users of this superb lens were.

At this point^*, Canon is the only manufacturer in its industry with two proprietary high-performance anti-reflective film technologies like SWC and ASC that totally differ in properties. Our development efforts continue towards making lenses with even higher performance through the clever application of these two lens coatings.

• * As of June 2018