Download Fundamentals of Optics: Refraction and Reflection of Light - Prof. Victor Vladimir Gurarie and more Study notes Physics in PDF only on Docsity!
Light and Optics: We just learned that light is a wave (an "electromagnetic wave", with very small wavelength). But in many cases, you can safely ignore the wave nature of light! Light was studied for a long time (obviously), long before Maxwell, and very well understood. People thought about light as sort of like a stream of "particles" that travel in straight lines (called " light rays "). Unlike particles, waves behave in funny ways - e.g. they bend around corners. (Think of sound coming through a doorway.) But, the smaller the λ is, the weaker these funny effects are, so for light ( tiny λ), no one noticed the "wave nature" at all, for a long time. λ of light is 100 x's smaller than the diameter of a human hair! For the rest of this term, we'll study the more "classical" aspects, called GEOMETRICAL OPTICS - the study of how light travels, and how we perceive and manipulate it with mirrors and lenses.
- We will ignore time oscillations/variations (10^14 Hz is too fast to notice, generally!)
- We'll assume light travels in straight lines (at 3E8 m/s, super fast)
- Light can then change directions in 3 main ways: i) Bouncing off objects = reflection ii) Entering objects (e.g. glass) and bending = refraction iii) Getting caught, and heating the object = absorption (iv) Bending around objects = diffraction is a subject for the next Chapter - which we won't have time to get into this semester! This is the place where the wave nature of light really comes into play. It does have applications - plenty - but you'll need to take another physics course, or study it on your own!)
How do you know where objects are? How do you see them? You deduce the location (distance and direction) in complicated physiological/ psychological ways, but it arises from the angle and intensity of the little "bundle" of light rays that make it into your eye. If light bounces off a smooth surface (like a mirror, or a lake), it's called " specular reflection ", and it is always true that Θ(i) = Θ(r), "angle of incidence"="angle of reflection" (See sketch for the precise definition of which angles those are!) If light bounces off a dull surface (like e.g. white paper, or a wall), it's called "diffuse reflection", and the light comes out every which way. (Microscopically, "dull" means that the surface is not smooth on the scale of the wavelength of light) (Of course, microscopically, the reflections are still specular, but the surface is wiggly so the net result is diffuse. By the way, the "angle of incidence = angle of reflection" rule from a smooth surface (e.g. a shiny metallic surface) arises from Maxwell's equations! ALL of the classical properties of light ultimately arise from these fundamental underlying equations!) (Only those in this "bundle" reach [and enter] your eye) (Many rays leave) "normal" light out ("reflected") light in ("incident") mirror θ i^ θ^ r out in in in in in out out out out out out in in^ in
Any transparent medium (air, H20, glass,...) that lets light through will have a number n, the " index of refraction ", associated with it. n is determined by how fast light travels through the material. (Light only travels at c, the "speed of light", in vacuum. In materials, it is always slowed down.) The bigger n, the slower the light travels: n = c/v = speed of light (in vacuum) / speed of light (in medium) = 3E8 m/s / v (in medium) n >1 always. (another way to say this: light never goes faster than c!) Examples: air: n= 1. water: n=1. glass: n=1. diamond: n=2.4 (light travels less than half the normal speed in diamond!) Over in the JILA labs, there are experiments with materials that have n (at one special frequency, anyway) about 1E7, so large that the speed of light is about as slow as a person on a bike! If light goes from one medium into another, it will (in general) bend, i.e. change it direction. This is called refraction. This fact is a property of waves. Nevertheless, the description of the effect doesn't need to refer to the wavelength, or wave nature, of light, so this topic still belongs in this geometrical optics chapter. The math involved is fairly straightforward (one equation to learn), and there are many important consequences/applications, ranging from simple eyeglass lenses, to fancy telescopes, to medical imaging equipment, optical fibers for phone lines, etc...
- Light is SLOWER on the right side, where n2 is large.
- θ1 is larger, θ2 is smaller: Light gets bent "towards the normal" as it goes from low index (like air) into higher index (like glass) (And vice versa = you can always reverse a ray diagram like this) There is a formula for the refraction of light (as shown in this figure) derivable from Maxwell's Equations, called Snell's Law: n1 sin(θ1) = n2 sin(θ2). (See fig for definitions of symbols) Example: You are looking into water at a little fish.
- Your eye draws "straight lines" (dashed, in the fig) and deduces that the fish is located at the image point in the figure.
- The image you see is virtual , the light rays do NOT physically pass through the spot where they appear to originate from. (If you placed a black card right in front of the spot where you perceive the fish to be, you will still see the fish, the rays don't pass through the card's location.)
- The image appears to be slightly less deep (you have to think about that one, it requires that you carefully draw more than one ray, because you can't figure out "depth" or "distance" with only one light ray.
- Light rays are always reversible. That means the fish perceives you in another place too - higher than you really are. (Can you see why?) Medium # n1 (small) (e.g. air) boundary line, or "interface" Medium # n2 > n (e.g. glass) normal to boundary line θ 2 θ 1 incident light transmitted light (air) (water) Image object
Example of total internal reflection: "optical fibers", used e.g. as an "endoscope" (a medical tool to look inside of bodies) You take a thin, flexible fiber of glass (or really a whole bundle of them), which can be threaded into a body. Light enters at one end inside the body, but the light rays are all bouncing at a shallow angle (θ greater than θcrit, do you see this?) against the glass/air boundary, and totally reflect, bouncing their way along never getting out till they reach the end, outside the body. Light rays can pass through several boundaries. For example, you might have a sheet of glass - a light ray will enter (going from small n1 to larger n2) and then exit (large n2 -> small n1) At each boundary , Snell's law will hold. At the left boundary we have n1 sin(θin) = n2 sin(θ2) (light bends toward the normal - convince yourself of the equation and the physics) At the right boundary we have n2 sin(θ3) = n1 sin(θout) (light bends away from the normal - again, convince yourself) But geometry tells us (if the walls are parallel) that θ2 = θ 3 (do you see why?) Which means sin(θ2) = sin(θ3). So n1 sin(θin) = n2 sin(θ2) = n2 sin(θ3) = n1 sin(θout) (can you follow all the steps required to write that last line down?) That means (compare the far left with the far right of that eqn) sin(θin) = sin(θout), which says θin = θout. Conclusion: the light ray is displaced sideways a tiny bit, but it is not bent, overall. (Glass over paintings doesn't distort the image like looking into a fishbowl does) glass or plastic air (small n2) ( big n2)
air
(n1)
glass
(n2)
θ in
incident
air
(n1)
θ out
What if you have glass with walls that are not parallel? The incoming ray is now bent. You have to think about this a bit, but it will always be bent away from the thinner part of the glass... This is the ideas behind lenses. As light enters, it is bent. and rays come out different depending on where and how they strike. The geometry looks complicated (and it is!) but for thin lenses, the result is relatively simple. First consider a bundle of parallel, horizontal rays entering a convex lens from the left, as shown below. The central ray sees two parallel edges, so it is not deflected at all. It goes straight on through. The rays striking the edges bend away from the thinnest part, as shown. This (convex) lens is called a "converging lens". All the parallel incoming rays are bent towards a common point, the focus, or focal point. (Of course, the incoming rays don't have to be parallel - in which case we'll have to think about what happens! We'll get to that soon) How could you produce parallel incoming rays like that? Lots of ways. Simplest example: having a small "pointlike" source very far away... Although the source sends out rays in all directions, the only rays that will make it to you are those that happened to be going in YOUR direction to start with. You will see only those rays, which are nearly parallel... (you need to think about that a little - draw a picture for yourself)
air (n1)
glass
(n2)
θ in
inc. air (n1)
θ out
convex lenses
(contact lens)
Parallel rays in
focal length,
f
focus
optical
axis
Optometrists define P = 1/f = "power of lens". (It's a poor name - it has nothing to do with power as in "watts". The units of P are m^-1, and are quoted as "diopters".) 1 D = 1 diopter = 1 m^-1. A "5 D lens" means f = 1/P = 1/5 m-1^ = 0.2 m, or 20 cm focal length. (We'll see what the consequences of this are, shortly) What if the object is not at “infinity”, so incoming rays are not parallel? ν Image is still formed ν No longer in “focal plane” ν May be real, or virtual: Depends on lens, and how close object is. The Lens Equation tells you all sorts of useful info: 1 f
do
di where do = "object distance" di = "image distance" f = "focal length". (The equation is proven in the text , if you are interested. It's basically just simple geometry. See the figure for definitions of symbols.) Comments and examples:
- if do=f (i.e. an object at the focus on the left), then 1/f = 1/f + 1/di. The only solution to that is di=∞ (think about it) We've seen the picture for this case before: (notes p.8 and 9) There is no image, or if you like, the image is “off at infinity”.
- If do = ∞ (i.e. if object is very far away) then 1/f = 1/∞ + 1/di which says di = f. We already knew this: image of faraway sources is in the focal plane, a distance f from the lens.
- If do = 2f, then we have 1/f = 1/(2f) + 1/di, solving for di gives di = 2f. (check for yourself) There is something pleasingly symmetric about this result: an object at twice the focal distance focuses at twice f on the other side. Draw a picture for yourself.
object
di
image
do
f
As do decreases from ∞, the formula says di increases (from f). Why? You can think about the math yourself, but the physics can be seen from sketches: Image is barely past f, the focal point. A given lens can only bend light so much. If the rays are coming in not quite parallel, as shown, the lens can't bend them as far "inward" as it did when they really came in exactly parallel. If the source is nearer still, the rays come into the lens at a yet steeper angle, the lens can’t bend them down to the original focus, it can’t bend that much! They will focus, but farther away. (That's what the formula gives too) The lens equation is great, but it is essential that you also be able to make a sketch to crudely estimate what’s going on! These sketches are called “ ray diagrams .” It also helps to see what’s going on when the source is “extended” rather than a point source. For you to think about: if the focal length is shorter (so the "power" P=1/f is larger), what happens to the location of the image, for a given object distance?
Source far, f
not infinite
do di
Source closer f
do di
In cameras the focal length f is generally fixed. (It just depends on the shape of the lens, and is not usually variable) Since (1/f) = (1/do) + (1/di), the “image distance” di depends on do. “di” is where the film needs to be, so you must adjust the distance of film depending on how far away the object is. (this is called “focusing the camera,” the lens <-> film distance changes) In eyeballs the distance of the retina behind the lens (“di”) is fixed by the size of your eye. So, your eye adjusts the shape of the lens, and thus effectively changes “f”! Camera examples: A “50 mm lens” means f = 50 mm = 0.05 m. An object at infinity means the film should be .05 m behind the lens. As object gets closer => film should be moved further back, away from lens. That's what's happening when you twist-focus an SLR camera. [note: “35mm Camera” refers to the film size, not lens size.] Eyeball examples: In general thin lens = less bending = larger f Fat lens =more bending= smaller f
- Object at infinity
- eye relaxed
- lens is thin
- f is the size of the eyeball In the figure to the right:
- Object is closer = need to bend light more sharply to get it to focus on retina
- need fatter lens (smaller f)
- Muscles squeeze the lens to do this.
- Can’t make f small enough, (can’t bend rays sharp enough) = can’t focus! See fuzzy image… Near point (“N”) = shortest distance you can still focus to. Typically, N~ 25 cm (~10 in) for normal (young) eyes. far object f nearer object ciliary muscles squeeze lens fatter object too close!
If you are “near sighted” ( myopic ) like me, you can see nearby things, but not far away objects. Your “relaxed” eye lens is still too fat, and bends light too much:
- Lens is too fat in relaxed state
- Focus point is inside the eyeball
- You see “fuzzy image” when object is far away.
- This (myopic) lens has a “far point”, which is the farthest point it can still focus on.
- Relaxed eye
- Can focus on this (nearer) object. The myopic eye is "overbending" the light. To correct this problem (like my glasses), you need a lens in front of the eye which should diverge the light a little bit. A diverging (concave) lens. Numerical Convention: “f” for a diverging lens is negative. Parallel light rays ( in) diverge out: They appear to be coming from focus on the back side, as shown. So, a far away object on the left will produce a virtual image: The light rays don’t really all pass through that point! far object myopia far pt.
Parallel rays in
f f
Summary of sign conventions: Converging lens f > 0 Diverging lens f < 0 Real object do > 0 Real image di > 0 Virtual image di < 0 Upright image hi > 0 Inverted image hi < 0 Example: Object located inside of focus of converging lens. Let f=5 cm, do = 2 cm (object is located 2 cm to left of lens) Ray #1 goes straight through. Ray #2 is parallel on left, then bends towards focus on right. Ray #3 is on the line from the focus on the left, comes out parallel. Picture says image is virtual, upright, enlarged, inside focus. Or, we can use the formula to get this as well: 1/f = 1/do + 1/di. Therefore 1/di = 1/f - 1/d0 = 1/(5 cm) - 1/(2 cm) = -0.3 cm^- So di = -3.3 cm. Since di is negative => virtual image. di < 5 cm => inside focus, m = -di/do = +3.3/2 = +1.7, sign says it is upright, fact that m>1 says it is enlarged. This is a magnifying glass! You hold it close to the object, and then you see a bigger image. It appears to be coming from the "far side" of the lens... it's virtual... object (on left) lens image di do (^) f ho hi Ray # Ray # Ray # f
If you want an object to look larger, you bring it closer (right?) But you can’t bring it closer than “N”, 'cause then it gets fuzzy/hard to see. The max angle θο an object of height ho can “span” inside your eye is thus tan θο (max)= h0/N (Look at the figure, convince yourself) But, what if θo(max) is still too small for you? Magnify θ with a lens! Magnifying Glass Same principle as the example on the previous page: Usually put object at focus, so rays come out parallel. Choose a glass with focal length f < N ~ 25 cm (typically) Object at focus => rays come out parallel. Now, put your eye in front of the lens (on the right.) Parallel rays come into your eye = you see the tip as a pointlike object at “infinity,” tipped up to an angle tan(θ') ≈ ho / f.
- You can always approximate tanθ ≈θ for small θ. (Remember that one? It's true only if θ is measured in radians though. ) So what we have here is Angular Magnification. The formula is M = θ'/θo(max) (this defines "angular magnification".) ≈ (h0/f) / (h0/N) = N/f. (from the geometry of the picture, making that small angle approximation) What's going on is this: the presence of the lens makes the object appear to be farther away (hence, easier to focus on) and also subtend a larger angle in your eye (which means it looks bigger) The ratio of the "new angle subtended with the lens" (θ') to the biggest angle you could get by bringing it close (θo(max)) without any lens, is the angular magnification. E.g: f (lens)=5 cm gives you (if you have a typical near point) M=N/f~25 cm/5 cm=“5X” (5 times angular magnification) (Note that "M" is different from "m = lateral magnification".) ho object θο N (or more) image do (=f) ho f f object θ '
