Working With Optics

Optical Hardware

Optomechanics, apart from being a very clever-sounding word, is the technical term for all the bits of hardware used to mount and manipulate optics. This section will walk you through some of the basic hardware you’ll use when working with optics. All images in this section come from Thorlabs, since that's mostly the equipment we have in the lab. First, here are a few of the more common pieces of hardware.

Untitled Database

!https://static.miraheze.org/cdiwiki/4/4d/Washers2.png

The force of a screw/washer is spread out over an area more than six times larger than that created by just a screw alone.

Screws and washers

Probably the most common mistake made with screws is forgetting to use a washer. A washer both spreads out the force of the screw head and presents a smooth, rounded surface. Without that barrier, the screw will gouge out a tiny imprint of itself where it was tightened. That’s fine if the screw is going in a screw-sized hole, but it becomes a problem when the screw is going into a long slot because it’s very difficult to tighten a screw into a nearby position without it slipping into the gouge. Any time you use a screw in a slot, there needs to be a washer (smooth side down) under the screw cap.

The other common mistake involves mixing up the two types of setscrew (see the image in the table). Steel-tipped setscrews are used to connect two tapped holes, such as the top of a post and the bottom of a lens mount. Plastic-tipped setscrews are used to exert a pressure against a flat surface, such as holding a mirror in its mount. A steel-tipped setscrew used in the wrong way can gouge metal (bad) and break glass (very bad). Never tighten a steel-tipped setscrew against an optic!

Pre-mounting

Before we talk about the mounts for the optics, we need to talk about the mounts for the mounts. To avoid confusion, I’ll refer to this as pre-mounting, but you should be aware that you won’t find that term anywhere else because I just made it up. The post holder (I’ll let you guess what it’s for) can be mounted to the table in a number of ways. You could screw a pedestal base onto the bottom and clamp it down with a fork. You could also attach a baseplate to the bottom and screw that onto the table. Simply dropping a screw through the post holder and screwing it into one of the holes on the table is not recommended because it severely limits your ability to adjust the position.

If you end up working with a lot of precision optical systems, you will develop a deep and abiding appreciation for whoever invented the translation stage. Basically, it lets you translate a small part of the optical table within a range of a few inches. With a single stage, you can move a lens to focus exactly where you want, or adjust the path length in one arm of an interferometer. With a combination of stages, you can build some pretty sophisticated positioning systems.

Optic mounts

In a very general sense, the term "optic" refers to something that interacts with your light, and is made mostly out of glass. The two basic types of optics are reflective optics (e.g. mirrors and beamsplitters) and refractive optics (e.g. lenses). It’s almost always best to put a mirror in a kinematic mount which allows for fine angular adjustment. However, a kinematic mount can sometimes get in the way, such as when used with a beamsplitter. In such cases, a fixed mount is a good alternative. In both cases, the optic is placed in the mount and fastened using a plastic-tipped setscrew. For a lens, you’ll use the aptly named lens mount, which fastens the optic with a retaining ring. All three types of mount are fastened (either with a setscrew or cap screw) on top of a post.

When mounting the optics themselves, its best to put the mount on a post first. If you need to move a mounted optic from one post to another, hold the mount still and rotate the post. Both of these practices just reduce the likelihood of accidentally touching the optical surface.

Alignment

Aligning optics requires practice and patience. Once you get the hang of it, though, you may find a kind of zen in the precise, methodical, perfection-seeking process. Or it may just be a necessary evil. Either way, if you’re going to be sticking around, you should probably know how to do it well.

Let’s start by defining a 3D coordinate system above the optical table. The x-axis is defined to be parallel to the table, perpendicular to the beam, with +x left of the beam. The y-axis is defined to be perpendicular to both the beam and the table, with +y pointed up from the table. The z-axis is defined to be parallel to both the beam and the table, with +z being in the direction of beam propagation.^[1]

The vertical angle of the beam (rotation about the x-axis) is called ϕ. The horizontal angle of the beam (rotation about the y-axis) is called θ. Defining the coordinates in this way is fairly common practice.

Controlling beam direction

!https://static.miraheze.org/cdiwiki/thumb/6/62/Dog-leg.png/400px-Dog-leg.png

A laser beam, like any 3-dimensional line, can be defined by six numbers. These might be two points (x1, y1, z1), (x2, y2, z2) or they might be a single point (x, y, z), and a direction (ϕ, θ, ±z). We’ll constrain these six degrees of freedom by setting up two apertures—two points that your beam has to pass through. Any aperture will do, but it’s best to use irises, since they can be opened and shut without changing position. We can then use a pair of mirrors to align the beam with those two irises, as shown in this image. Align first mirror with the first iris, then open the first iris and align the second mirror with the second iris. Then just alternate back and forth until the beam passes nicely through both apertures.^[2] This two-mirror-two-iris arrangement is sometimes called an optical "Z" or a dog-leg. There are many variations on this method. For example, the exact placement of the mirrors is somewhat irrelevant. The setup shown here can take up a lot of space; more compact arrangements are depicted here. Another variation is to fix two post holders in place and move a single iris back and forth between the two. By placing a clamp on the iris post, you can ensure that the beamline is parallel to the table (i.e., the aperture is the same height at both locations). All these variations, however, come back to the concept of two mirrors aligning the beam to go through two points.

Controlling beam size

!https://static.miraheze.org/cdiwiki/thumb/6/6c/Telescope.png/400px-Telescope.png

Now that we know exactly where our beam is going, let’s make sure it’s the right size. This is accomplished by arranging two lenses (focal lengths f1 and f2) in an optical telescope, as shown. The two lenses are separated by a distance d=f1+f2, which allows them to share a common focal point. This, in turn, ensures that a collimated beam going into the telescope will still be collimated coming out the other side. The optical telescope will magnify the beam diameter by a factor of M=−f2/f1.^[3] This means that placing the longer focal length first will shrink the beam, whereas placing the shorter focal length first will expand the beam. A note on aligning the telescope. Always align your lenses one at a time, starting with the longer^[4] focal length. For each lens, ensure that the beam comes back to the same spot and that the lens is aligned with the optical axis (i.e. ϕlens=θlens=0).

Lens abberations

If you’ve taken Physics 123, you probably know about spherical aberration. Light hitting the edges of a spherical lens focuses to a different point than light hitting the center. This causes the focus to appear blurred. There are several other aberrations you should be aware of. Coma occurs when the optical axis of the lens is not aligned with the optical axis of the beam, and causes the focus to appear wedge-shaped. Coma is one of the worst aberrations, but also the most easily prevented—you only encounter it with improperly aligned lenses. Astigmatism occurs when a lens is not rotationally symmetrical, and therefore focuses light differently along different axes, causing the focus to appear as a line. Petzval field curvature is related to spherical aberration, and occurs when the edges of an image cannot be brought into focus in the same plane as the center of the image; this won’t affect the focus of a beam, but it may affect the edges of diffraction patterns in the Fourier plane. Chromatic aberration occurs when a lens focuses different wavelengths to different points, causing a color smear.

!https://static.miraheze.org/cdiwiki/thumb/7/7f/Flat2curve.png/400px-Flat2curve.png

The flat-to-curved rule for plano-convex lenses.

The following techniques will help reduce lens aberrations.

• Use the rule "flat-to-curved, curved-to-flat" for plano-convex lenses. Flat wavefronts (i.e. a collimated beam) should hit the curved side of the lens, and curved wavefronts (i.e. a diverging or converging beam) should hit the flat side of the lens. This reduces spherical aberration.
• Aim for the center of the lens. Nearly all aberrations are worse at the edges of a lens than in the center.
• Never settle for a misaligned lens. Just because the spot goes where you want it to go, that doesn’t mean that your lens is aligned. Make sure that the spot is in the right place, AND the lens is parallel with the beam.

Care and cleaning

The word "precision" sometimes gets applied to things like metal-cased mechanical pencils or fancy cake-decorating kits. Precision optics make these look like the stone tools of the Neanderthal. If you were to take one of the basic little one-inch optical mirrors we have in the lab and scale it up to the size of BYU campus, the highest point on the mirror would be about 2 cm higher than the lowest point. Optical precision is simply astounding.

This makes cleaning optics... tricky. You see, when you clean your bathroom mirror at home, you’re not worried about how many Windex molecules you’re leaving the surface, or whether you’re making nanometer-deep scratches in the glass. Such imperfections are simply too small to worry about. But when you’re dealing with precision optics, it’s a whole different story. With that dramatic introduction, here are some good practices for handling and cleaning optics, taken from more extensive tutorials published by Newport, Edmund Optics, and Thorlabs.

Proper handling

• Always wear gloves. Your skin is literally covered in corrosive oils (no offense). Gloves prevent those oils from coming in contact with the optic.
• NEVER touch the optical surface(s). Even with gloves, you should always handle an optic by the edges.
• NEVER place optics on hard surfaces. With no cushion on either side, any contaminant between the optic and the surface will be ground in. This is especially true for convex lenses.
• Store optics by wrapping in a clean lens tissue, then placing in a properly sized optic storage box. The one exception to this is single-sided mirrors, which can be placed face-down in most optic storage boxes without a lens tissue.
• NEVER blow on an optic. Even the most careful breath has some saliva in it, which can damage the optic.
• Clean dirty optics promptly. Many contaminants do more damage the longer they sit.

Proper cleaning

• Keep the optic clean. What’s the best way to clean an optic? Don’t.
• Dust off the optic with compressed air. This should ALWAYS be the first step when cleaning. Keep the can upright and use short bursts to prevent any of the compression liquid from contacting the optic. If there are no visible stains after dusting, you don’t need to go any further.
• Any cleaning beyond compressed air should be supervised for the first few times. Additionally, some optics (such as ruled-grating mirrors) should NEVER be touched, even to clean them.
• Treat lens tissues with as much care as the optics themselves. If it has touched anything besides the inside of its packaging, it’s considered dirty. Make sure parts of the tissue that you handle don’t touch the optic. NEVER reuse a lens tissue.
• Use the drip-drag method for mirrors and other flat optics. Place a clean lens tissue over the optic. Place a single drop of solvent on the tissue so that it soaks through and sticks to the optic. Drag the tissue steadily across the optic until the solvent is all gone.
• Use the brush method for lenses and other curved optical surfaces. Fold the lens tissue until it’s about the size of your optic, making sure not to touch the part that will become the brush. Clamp the folded tissue with a hemostat, and wet it with solvent. Shake off excess solvent, and wipe ONCE across the surface of your optic.

Steps for beginners

If you're working through the steps for beginners, follow these instructions. They are a bit vague by design, because figuring things out yourself is far better than simply following a recipe. At each step, you should take some time to (safely) play with it and gain an intuition for how things work.

Beam alignment

Quick safety reminder! There are many reflective surfaces in the lab, and not all of them are obvious. For example, in a moment you'll be pointing the laser in the general direction of the sink. If the beam hits any part of the faucet, it will be reflected in ways that are practically impossible to predict. It's a good idea to keep the laser powered down when possible, such as while you're mounting it. Once you turn it on, keep an eye out for stray beams, especially if they're pointed toward the doors. If you do notice a stray beam, block it with something non-reflective (and preferably dark-colored).

Mount the laser on the table so that it is pointed toward the far wall (where the sink is).

Place the variable attenuator so that it reflects the beam back next to (but not directly back into) the laser emitter to prevent a stray beam.

• Why wouldn't you want the reflection to go directly back into the laser? Hint: A laser cavity is basically just two mirrors facing each other.

Using two mirrors and two irises as shown above, align the beam along a line of holes in the table. Put the irises at least two feet apart, since some optics will need to go between them.

• Which types of mounts would be best for getting the irises to line up with the table grid?

Pinhole diffraction

Once the beam is aligned, place a pinhole just after the first iris. A translation stage may be helpful here.

• How can you tell if the pinhole is centered on the beam? Hint: Where is the beam brightest?

Using a sheet of paper or cardboard, examine the diffraction pattern formed by the pinhole.

• Does the diffraction pattern look like an Airy disk?
• How can you tell that the beam is no longer collimated after passing through the pinhole?
• Should a larger pinhole create a larger or smaller diffraction pattern? Hint: What kind of diffraction pattern would be made by an infinitely large pinhole?

Now place a lens behind the pinhole to re-collimate the beam. Be sure to align the lens so that the beam still goes through the second pinhole.

• How can you tell when the beam is collimated?
• All of the lenses in our lab are plano-convex. Which side should face the pinhole?
• How far from the pinhole does the lens need to be to collimate the beam?
• Is this distance the same for different lenses?

Image detection

If you want to use your own computer, you will need to install the Mightex drivers and software (Windows only) as well as an image analysis software such as ImageJ. Otherwise, everything you need is already on the lab computer.

Mount the Mightex camera behind the second iris. Using the SSClassicCameraApp software to see where the beam hits, align the camera so that the diffraction pattern is centered.

• Once the camera is aligned, is the second pinhole still necessary? Why or why not? Hint: An argument can be made for both answers, so give it some thought.

Once the camera is in place, open up the software. If the camera is plugged in and the correct drivers are installed, you should see the camera's name and an option to open it in 8-bit or 16-bit mode. Select 16-bit and click "OK" to go to the main interface.^[5] Make sure that the following settings are correct:

The arrow button in the top left turns on the camera, the blue rectangle button in the top right lets you view the camera live, and the 90's-looking camcorder button in the file control section saves frames.
Saturation describes when an instrument is unable to faithfully measure signals above a certain intensity. This is often more nuanced than simply a hard cutoff. For example, most digital cameras begin to lose linearity at around 90% of their maximum value, meaning that a 1% increase in light levels no longer corresponds to a 1% increase in camera response. Therefore, a good rule of thumb is to keep the brightest part of your image at about 85% of the camera's maximum. In general, you can do this by adjusting either the beam intensity or the exposure time. Set the exposure as short as it will go, adjust the beam intensity accordingly, and save an image with the exposure time in the file name. Then set the exposure to a half-second (you may need to turn off the lights), adjust the beam intensity again, and save another image. Try several other combinations of intensity/exposure.

• Before analyzing the images, try to think of one advantage and one disadvantage of each extreme (high-intensity-short-exposure and low-intensity-long-exposure).

Image analysis

Now pull up the images in ImageJ. Play around with some of the tools in the Image, Process, and Analyze menus and see what you can learn from your measurements.

• Use Image > Adjust > Brightness & Contrast... to reduce the maximum pixel value. What do you notice about the low-intensity features?
• After analyzing the images, do they match what you expected? Try to think of at least one more advantage/disadvantage of each method
• What is the diameter of the first dark ring? Hint: Detailed camera specifications like pixel size can be found online.

One way to reduce the impact of intermittent noise is by summing several measurements together. If the noise is truly random from frame to frame, it will average out. Make a new folder and set it as the Mightex saving location. Then set "Grab Frames" to 20, and take the images. Open them as a stack in ImageJ by dragging the whole folder onto the little ImageJ window. If it pops up with a bunch of questions, just click ok. Now go to Image > Stacks > Z Project... and do "sum slices."

• Zoom into a bright region of both the raw stack and the summed image. How are they different?
• Reduce the maximum again to look at the low-intensity features in both images. How are they different?

Take another 20 images with the laser turned off to get a dark-field image. Sum the background images and subtract the result from the diffraction image (Process > Image Calculator...)

• Is the background uniform? Should it be?
• Are there some artifacts that persist despite background subtraction? What might cause these?

Pick your favorite image and examine the effects of downsizing with Image > Adjust > Size...

• How can you tell if the diffraction pattern is Nyquist sampled?

Notes

1. ↑ This can be a bit confusing when we start to redirect the beam with mirrors. Generally, you will define z^ based on the region of the beam you’re dealing with. There’s no standard way to handle a region that includes a reflection or other directional change; use whatever makes the most sense given the context.
2. ↑ If you align the second mirror with the first aperture and vice versa, the beam will actually diverge away from the line you’ve defined.
3. ↑ The negative in that factor just means rays that hit the bottom of the first lens will hit the top of the second lens, and vice versa. It’s common to just drop the negative in cases where the beam is symmetrical.
4. ↑ The order isn’t strictly necessary, but starting with the tight lens tends to make the spot impractically large for precise alignment.
5. ↑ At this point, you may notice is that the user interface isn't as pretty as, say, your iPhone camera app. That's fairly common in science—function comes before form.