An optical flat is an optical-grade piece of glass lapped and polished to be extremely flat on one or both sides, usually within a few millionths of an inch (about 25 nanometres). They are used with a monochromatic light to determine the flatness of other optical surfaces by interference.[1] When an optical flat is placed on another surface and illuminated, the light waves reflect off both the bottom surface of the flat and the surface it is resting on. This causes a phenomenon similar to thin-film interference. The reflected waves interfere, creating a pattern of interference fringes visible as light and dark bands. The spacing between the fringes is smaller where the gap is changing more rapidly, indicating a departure from flatness in one of the two surfaces, in a similar way to the contour lines on a map. A flat surface is indicated by a pattern of straight, parallel fringes with equal spacing, while other patterns indicate uneven surfaces. Two adjacent fringes indicate a difference in elevation of one-half wavelength of the light used, so by counting the fringes differences in elevation of the surface can be measured to millionths of an inch.
Usually only one of the two surfaces is made optically flat to the specified tolerance, and this surface is indicated by an arrow on the edge of the glass.
Optical flats are sometimes given an optical coating and used as precision mirrors for special purposes, such as in a Fabry–Pérot interferometer or laser cavity. Optical flats have uses in spectrophotometry as well.
Flatness testing
An optical flat is usually placed upon a flat surface to be tested. If the surface is clean and reflective enough, rainbow colored bands of interference fringes will form when the test piece is illuminated with white light. However, if a monochromatic light is used to illuminate the work piece, such as helium, low-pressure sodium, or a laser, then a series of dark and light interference fringes will form. These interference fringes determine the flatness of the work piece, relative to the optical flat, to within a fraction of the wavelength of the light. If both surfaces are perfectly the same flatness and parallel to each other, no interference fringes will form. However, there is usually some air trapped between the surfaces. If the surfaces are flat, but a tiny optical wedge of air exists between them, then straight, parallel interference fringes will form, indicating the angle of the wedge (ie: more, thinner fringes indicate a steeper wedge while fewer but wider fringes indicate less of a wedge). The shape of the fringes also indicate the shape of the test surface, because fringes with a bend, a contour, or rings indicate high and low points on the surface, such as rounded edges, hills or valleys, or convex and concave surfaces.
Preparation
Both the optical flat and the surface to be tested need to be extremely clean. The tiniest bit of dust settling between the surfaces can ruin the results. Even the thickness of a streak or a fingerprint on the surfaces can be enough to change the width of the gap between them. Before the test, the surfaces are usually cleaned very thoroughly. Most commonly, acetone is used as the cleaning agent, because it dissolves most oils and it evaporates completely, leaving no residue. Typically, the surface will be cleaned using the "drag" method, in which a lint-free, scratch-free tissue is wetted, stretched, and dragged across the surface, pulling any impurities along with it. This process is usually performed dozens of times, ensuring that the surface is completely free of impurities. A new tissue will need to be used each time, to prevent recontamination of the surfaces from previously removed dust and oils.Testing is often done in a clean-room or another dust-free environment, keeping the dust from settling on the surfaces between cleaning and assembly. Sometimes, the surfaces may be assembled by sliding them together, helping to scrape off any dust that might happen to land on the flat. The testing is usually done in a temperature-controlled environment to prevent any distortions in the glass, and needs to be performed on a very stable work-surface. After testing, the flats are usually cleaned again and stored in a protective case, and are often kept in a temperature-controlled environment until used again.
Lighting
For the best test-results, a monochromatic light, consisting of only a single wavelength, is used to illuminate the flats. To show the fringes properly, several factors need to be taken into account when setting up the light source, such as the angle of incidence between the light and the observer, the angular size size of the light source in relation to the pupil of the eye, and the homogeneity of the light source when reflected off of the glass.
Many sources for monochromatic light can be used. Most lasers emit light of a very narrow bandwidth, and often provide a suitable light source. A helium-neon laser emits light at 632 nanometers (red), while a frequency doubled Nd:YAG laser enits light at 532 nm (green). Various laser diodes emit light in red, green or blue. Dye lasers can be tuned to emit nearly any color. However, lasers also experience a phenomenon called laser speckle, which shows up in the fringes.
Several gas or metal-vapor lamps can also be used. When operated at low pressure and current, these lamps generally produce light in various spectral lines, with one or two lines being most predominant. Because these lines are very narrow, the lamps can be combined with narrow-bandwidth filters to isolate the strongest line. A helium-discharge lamp will produce a line at 587.6 nm (yellow) while a mercury-vapor lamp produces a line at 546.1 (yellowish green). Cadmium vapor produces a line at 643.8 (red), but low pressure sodium produces a line at 589.3 (yellow). Of all the lights, low pressure sodium is the only one that produces a single line, requiring no filter.
The fringes only appear in the reflection of the light source, so the optical flat must be viewed from the exact angle of incidence that the light shines upon it. If viewed from a zero degree angle (from directly above), the light must also be at a zero degree angle. As the viewing angle changes, the lighting angle must also change. The light must be positioned so that its reflection can be seen covering the entire surface. Also, the angular size of the light source needs to be many times greater than the eye. For example, if an incandescent light is used, the fringes may only show up in the reflection of the filament. By moving the lamp much closer to the flat, the angular size becomes larger and the filament may appear to cover the entire flat, giving clearer readings. Sometimes, a diffuser may be used, such as the powder coating inside frosted bulbs, to provide a homogenous reflection off the glass. Typically, the measurements will be more accurate when the light source is as close to the flat as possible, but the eye is as far away as possible.
Wringing
Wringing occurs when nearly all of the air becomes forced out from between the surfaces, causing the surfaces to lock together through the vacuum between them. The flatter the surfaces; the better they will wring together, especially when the flatness extends all the way to the edges. If two surfaces are very flat, they may become wrung together so tightly that a lot of force may be needed to separate them.
The interference fringes typically only form once the optical flat begins to wring to the testing surface. If the surfaces are clean and very flat, they will begin to wring almost immediately after the first contact. After wringing begins, as air is slowly forced out from between the surfaces, an optical wedge forms between the surfaces. The interference fringes form perpendicular to this wedge. As the air is forced out, the fringes will appear to move toward the thickest gap, spreading out and becoming wider but fewer. As the air is forced out, the vacuum holding the surfaces together becomes stronger. The optical flat should usually never be allowed to fully wring to the surface, otherwise it can be scratched or even broken when separating them. In some cases, if left for many hours, a block of wood may be needed to knock them loose. Testing flatness with an optical flat is typically done as soon a viable interference pattern develops, and then the surfaces are separated before they can fully wring. Because the angle of the wedge is extremely shallow and the gap extremely small, wringing may take a few hours to complete. Sliding the flat in relation to the surface can speed up wringing, but trying to press the air out will have little effect.
If the surfaces are insufficiently flat, if any oil films or impurities exist on the surface, or if slight dust-particles land between the surfaces, they may not wring at all. Therefore, the surfaces must be very clean and free of debris to get an accurate measurement.