by I. R. DaggSUMMARY, (Summary diagram on back cover) The translation matrix | + |. the refraction matrix | + a ou acat on aR oat The matrix connecting the planes through ay yy the vertices 0 and 0' of the optical system is 00" F is the first focal point and OF = -a),/a),. F? is the second focal point and O'F! = -ay,/ay)- The first focal length £ = -|a| /ay,- The second focal length f= -l/ay, + H is the first principal point TH = -f. For the negative FH(-) = f. H is the second principal point F'H' = ~£'. principal points: F'l'(-) = ¢", N is the first nodal point FN For the negative FN(-) = f". N' is the second nodal point F'N' = nodal points: FIN) = £ An image at point I' on the optic axis is formed of an object at point I on the optic axis and the magnification is m: FL=x xx! = ff" m= -f/x = -x'/f* £ l Wr = 6 3 NE=Vv fog +f ni wiee-w The rays of light are considered to proceed from left to right. A distance is considered positive in the unprimed system if measured from the reference point to the left, and positive in the primed system if measured from the reference point to the right.MATRIX OPTECS by I, R. Dagg, Ph.D., Professor of Physics. Appendix by J. Vanderkooy, Ph.D., Assistant Professor of Physics. University of Waterloo, Waterloo, Ontario, Canada. Otter Press, Box 747, Waterloo, Ontario.CHAPTER CONTENTS PAGE Introduction.....scccsssseeeessseeeeeesseeeee 3 Physical Basis of Geometrical Optics... 5 Basic Matrix Algebras.eeeeeeeee 6 The Basic Matrix Transformations of Geometrical Optics...... 11 The General Optical System. : seeeeseeeee 2D Image Formation...... sete teeeeee . 30 Practical Optical Systems.sssssessesseeeseeees 244 Spherical Mirrors. : veseeee 55 Appendix: Generalization of Matrix Optics for Arbitrary Ray Directions. seteeeeeees 63INTRODUCTION This book represents an alternative approach to that found in most standard texts to the study of image formation by spherical surfaces. As such it complements the usual treatment and may well be used along with another text. The text "Fundamentals of Optics” by F.A. Jenkins and H.E. White (Toronto, McGraw-Hill, 3rd ed., 1957) has been referred to throughout as it uses a sign convention which is consistent with that used in this book. The subject of geometrical optics starts with the laws of refraction and reflection for transparent media. It is then a question of using these laws to discover the properties of various optical systems which may contain any number of curved refracting and reflec~ ting surfaces. In principle we could solve any optical problem by the exact application of the basic laws. However, it is possible to derive some very useful general results by using approximate forms of the laws which treat only rays, called paraxial rays, which make omall angles with respect to an optic axis. Also, it is usual to confine the treatment to spherical surfaces. This approximate form of the basic laws as applied to paraxial rays in an optical system consisting of spherical surfaces is called the first order theory and constitutes a major part of the subject of geometrical optics. ‘The first order theory forms the basis for obtaining the equation describing image formation which relates image position to object position and defines magnification for the case of a single spherical refracting surface between two transparent media. For an extensive optical system this equation can be applied repeatedly toeach surface of the system in turn. However the equations so derived quickly become involved algebraically for all but the simplest cases, such as, for example, che single thin lens in air. The matrix method 1s another approach to the study of optical systens using the first order theory, By properly specifying the rays passing through the optical system it is possible to reduce a problem to a succession of matrix operations. This not only simplifies the algebraic manipulation but more importantly reveals some very general properties of optical systems not easily obtainable in other ways. From the mathematical point of view, the first order theory can be treated in terms of linear transformations. And, since matrix algebra is the most suitable vay of handling Linear transformations, it is obvious that we should apply the results of matrix algebra to treat the first order optical problem. The matrix method then represents a basic approach to the analysis of optical systems which results in discovering the cardinal points of a lens system from a more general point of view. The familiar lens equations for an optical system also result in both the Newtonian and Gaussian forms. This much is accomplished in 38 pages, part way through chapter 5. Included in this 1s a chapter which discusses the properties of two by two matrices as they describe the linear transformation involved in an optical system, The rest of chapter 5 shows a number of alternative ways of expressing image formation in terms of different cardinal points of a lens system. In chapter 6 some practical optical systems are discussed in terms of the matrix approach, Chapter 7 is a discussion of spherical mirrors by the matrix method. In the appendix it is shown that two by two matrices are sufficient for the treatment of the optical system in first order theory instead of the four by four matrices which describe non co-planar general ray directions.Physical Basis of Geometrical Optics As mentioned, the subject of geometrical optics is based on che laws of reflection and refraction. The laws of reflection state: 1, The reflected ray Lies in the plane of incidence. The angle of reflection equals the angle of incidence. The plane of incidence is the plane containing the ineident ray and the normal to the reflecting surface constructed at the point where the incident ray strikes the surface. the angle of reflection is the angle between the normal and the reflected ray; the angle of incidence is the angle between the normal and the incident ray. The laws of refraction state that for a refracting surface between two transparent media: The refracted (or transmitted) ray lies in the plane of incidence, and 2. nsin =n! sin 9", (Snell's law) , where ¢ is the angle of incidence and ¢' is the angle of refraction, the angle between the normal to the surface and the refracted ray, and a and n! are the indices of refraction of the media on either side of the refracting surface. The index of refraction of a medium is defined as the ratio of the speed of Light in vacuum to the speed of light in che medium, In first order theory, which essentially involves the first term in the series expansion for the sine of the angle, Snell's law is used in the form ng = n'p'. A better approximation than the first order theory involves higher order terms in the expansion which gives rise to the third and Fifeh order theories. The results of the higher order theories are usually expressed in terms of the defects in the image formed by spherical refracting surfaces, the so-called monochromatic aberrations. In what follows we shall concentrate on the subject of refraction, on the basis of first order theory, leaving the subject of reflection to be treated later as a special case.chapter 2 Linear Transformations and Matrix Menipulation The geometrical laws in terms of the first order theory may be stated in the form of linear transformations which in turn are most easily treated by matrix algebra, In particular we may represent the physical situation in terms of two linear equations. The two variables we shall call y and a. The physical meaning attached to these two variables will be stated later. We will be able to show that two other variables y' and a’ are related to y and a through two linear equations of the form: ayyta,a, o at = ay + ayo @) Here, a))5 4,55 83)» and ay, are constants, which we will later show to be characteristic of a given optical system. These constants may be written as the elements of a 2 X 2 matrix, @) In fact, equations (1) and (2) constitute a definition of equation (3). Now suppose that a further linear transformation is known which connects the variables y" and a" say, to y' and a' such that vw, 4) = by yi td 12 a2 a" . ()or in matrix notation a by. at (6) We can discover the equations which relate y and a to y" and a" by substituting equations (1) and (2) into (4) and (5). Then we have a" = by (ayy tay @ + dy, (ay ¥ tayo, a" = by, (ayy ¥ + ayy @) + by, (ayy ¥ + ayy wy or Oy ayy Pap ayy) y+ Oy ay typ gyda a aN = (boy ayy + bay Ag ¥ + (by1 ay + By Ag9) a @) which is a new transformation of the form yo ey yteya s (9) ey yt eye + (10) In matrix notation, y “U1 “12 y a’ Soy L091 al, qa) where the "cl coefficients are defined by comparing equations (9) and (10) with equations (7) and (8). The matrix formulation involves the substitution of the expression for the fyi column saerix[%] o¢ equation (3) into equation (6) to give 11) a 6 i 2 y ». al a2) 22| 81 822 ‘Thus comparing equations (11) and (12) we haveee bb aoa aa “ae nae wa ae ee | |b b aa : 43 2.22 an 22, aL 22 a) This essentially defines the usual rule of matrix multiplication given by the coefficients of equation (7) and (8) as: co sb a +b ay c= bo a +b a, n a Aa ra Man ae a1 Aha 12 822 c = b a +b ay, cob al +b ay aa ay “aa 22 Sad 22 21 “he 22 M22 or more compactly, aay jaa ‘The matrix equation may be written as Exercises 30.27 a aft 6} 6 45 ae on [3 2. show that [6 5] fo 3], [29 23] 2-1 12) 1. 4fe 3. Show that in general [A] [8] # [8] [A]- {c] = [B] [A]. 4. For vhat particular matrix 8 doos [A] (8) = [8] (al? sa] pa Ls 3 s. snow shat [53] [5 i] - [$ SF. 17} 6 3) _ far 17 o show ame [5 }(S3] - [Sy . - Lx wae 7. Show that the multiplication by ene matrix |} *] and any aacrie| always leaves the element a, unchanged. : towne [3] BY BE Determinant of a Matrix We shall have occasion to make use of one important property of matrices which involves the determinant of a matrix. ‘The determinant of the matrix [A] is written aslal ul 12 and is defined as the number (a a - a a) wy M22 Ba ae ‘The theorem that we are interested in states that the determinant of the product of a number of matrices is equal to the product of the determinants of each of the matrices forming the product. We leave the demonstration of this theorem to the exercises below. Exercises 6 af ]4s 30 27 fF, 9. show enat | $3 [22 | 1 3 | = 126 &5/)6 3], ]29 2}. 10. show that | $3] | $3 2 | =- v6 LL. Show that if (¢] = {8} (A] then [a] [al = [a] [a] = [cl. 12, Extend the result of exercise 11 to show that for N matrices such that (P] = (a) [a2] IP] = lar] lac! ta} Al. 13. Show that the result of exercise 12 holds for the product matrix of exercise 8 above. We are not going to be concerned with other properties of matrices. ‘The interested student may find these in any standard text on matrix algebra.= 10 - Figure 1 Figure 2-u- Chapter 3 ‘The Basic Matrix Transformations of Geometrical Optics Tn this chapter we shall first define the variables y and a that were used in the preceding pages. Then we shall derive two different Linear transformations, one corresponding to a translation, the other to a refraction operation. They will be characterized by a translation matrix and a refraction aatrix respectively. With these two matrices we shall be able to treat any optical system according to the first order theory. L, ‘The Variables Let us consider a ray of Light EF as in Fig. 1 which is traversing a uniform portion of a transparent medium which forms part of an optical system. The ray makes a small angle with respect to the X direction which we will designate as the optic axis of the system. The centres of curvature of all the spherical refracting surfaces lie on the optic axis. We shall choose a plane PP which is perpendicular to the optic axis. ‘The meaning of the variables y and a can be seen from Fig. 1. For a given plane PP the variable y represents the height sbove the X axis that the ray strikes the plane PP. The variable « represents the angle the ray makes with the X axis and since the slope SZ = tan a = a for small angles, we may take 4 as representing the slope of the ray as it passes the plane PP, The distance y above the plane 1s positive, below the plane negative. The sign of the angle a is determined from the sign of the slope and is positive in Fig. 1. ‘he Translation Matrix In Fig. 2 the ray EF is progressing (without change in direction) through the uniform medium as described for Fig. 1. We shall always draw-2- the ray as progressing from left to right in the X direction. We may designate two planes, one PP and the other QQ, a distance D from PP; the ray is characterised by the variables y, a,at PP and y$ a} at QQ. From Fig. 2 we may write y!=y +Da, (for small angles) and since a! equals a we may write another linear equation as als oyta. ‘These linear transformation equations can be written as ro] fy oat La as [4] if strikes a further plane WW at a distance T from QQ at angle «" and height y". In matrix form we could write 4 < 1 +m] [y © That is, y+(T+D a, a" =0+a, which can be verified by examining Fig. 3. The distances T and D are always taken as positive quantities as the transformation is defined in terms of the ray direction, that is,y" and a" are expressed in terms of y’ and a’ and y' and a’ in terms of y and a, The variables of a given plane are expressed in terms of the variables of the planes to the left of it.Figure 3-l4e 3. The Refraction Matrix At each surface separating two transparent media within an optical system, the direction of the ray may be changed according to Snell's law, which we take in the approximate form for small angles. We may express this refraction at the surface in terms of two linear equations which gives rise to the refraction matrix. In Fig. 4, VW represents the spherical refracting surface separating transparent media of indices n and n' respectively. The intersection of the surface and the optic axis is the vertex 0. The radius of curvature is R and is considered positive (as in the figure) if the direction 0¢ is the positive X direction where C is the centre of curvature of the surface. Since we have confined the discussion to small angles and rays close to the axis, the variables y and a may be considered to refer to the plane through 0 perpendicular to the optic axis. OPTIC AXTS-15- At the surface the angle a changes froma to a' while the variable y is simply the height above the X axis of the point at which the ray strikes the surface and refraction takes place. Thus we can write either y or y' for this height and the transformation equation may be written as yluy+0a . a6) We may now derive the transformation equation for « at this surface using Snell's law. From Fig. 4, n 6 = n'6! (approximately) or on (Sta) =n! (54a!) or nla! = nat(nen'") 6 and 8 = y/R. Therefore we obtain n'a! = na + (n-n') y/R. an ‘Thus the transformation equations can be written for the plane through 0 using equation (16) and rearranging (17) as, y+, (a6) “fale " or in matrix form a) y at] font a « as) nk ont and the matrix, a) aR on! is defined as the refraction matrix for the plane through the vertex 0 of the refracting surface.- 16 - ‘Translation and Refraction Now we have two basic matrix operations which describe the progress of a ray through an optical system. We need only apply the laws of matrix multipli- cation to determine the transformation between any two planes using the translation and refraction matrices, We illustrate this with the following example: A glass rod of index n' = 1.5 has a hemispherical convex surface of radius 6 cm on one end as in Fig. 5. A ray makes an angle « equal to 0.1 radians with the plane PP at a height of 1.6 cm from the optic axis. The plane PP is 8 cm to the left of the vertex 0. Find the height of the ray y' and its direction a! at the plane QQ 9 cm to the right of 0. Solution: Figure 5-u- x [yx] [3 0 Pia 2 “| 18 9 pq ok Note cince 1 Bow], fre -2 2] 3 fo a 18 9 |PQ 1 Now 1 | 4 -a 2 we Thus the solution is 1.8 cm and a! ~.0667 radians. _ r Bie 1.8 -.0667) + It is perhaps well to emphasize at this point that we have not yet mentioned the idea of image formation and y' of the example is certainly not the image of y, at a height y' which depends on a. since a ray originating at y will go through the plane 0Q If a were .05 radians say, in the last example then a ray through y = 1.6 cm passes the plane QQ at the height y'=1.3 cm instead of 1.8 cm. This is illustrated in Fig. 6. Figure 6- 18 - In anticipation of further sections, we can observe that I' in Fig. 5 appears to be an image of I as the ray along the optic axis (which is un~ deviated) and the ray at angle a both originate at I and pass through 1". To prove that I' is indeed an image of I we need to show that all the rays from I pass through I' regardless of the angle a as in Fig. 7. We choose the planes through I and I' as our reference planes and illustrate this with the help of the following exercise. ‘The numbers in this example have been chosen so that I' is 36cm to the tight of 0 and I is 24 cm to the left of 0. Find the matrix connecting the planes through I and I' and show that [*: [). T rt where now y', a', represents the ray at the I' plane and y, a, the ray at the I planes (s) aS sy Figure 7-19- In this exercise we note that y' = -y independent of a so that if y = 0 (point I) then y' = 0 as illustrated in Fig. 7 which shows I' to be an image of T. As can be seen from the above exercise, if the matrix element in the upper right hand corner (a,,) is zero, then the two planes are such that y' depends only on y. With this fact we can derive the usual equation (equation (21) below) for image formation by refraction at a spherical surface. Exercise 2 Let S be the distance OI and S' be OI’. If the index on the left is a, on the right n' and R is the radius of curvature, then s (ay if I' is an image of I. ‘ign Convention Equation (21) does presuppose the sign convention of Jenkins and white for example,in which the object distance § is positive if the object is to the left of 0 and the image distance S' is positive if the image is to the right of 0. In the above example we have used positive image and object distances. In Figs. 8 and 9 we illustrate negative distances which in these cases represent a virtual object and a virtual image. Generally we shall be able to extend the sign convention to a complex optical system in which the relevant distances are measured from various cardinal points which we shall define for the system. The cardinal points are readily obtained using the matrix approach.= 20- J siogative object Déacence Figure 8 egative Taage Discance >] Figure 9-2- chapter 4 The General Optical systen Ts Gonoral atric In the preceding discussion we vere able to describe the passage of a cay fron one point to another in a elaple optical systen by weans of two eraelation matrices and one refraction matrix. A gore complicated optical system nay contain many refeacting suc faces of different indices of refraction. Let us for example consider the eyaten of Fig. 10, A ray passes the plane #P at height y, angle 8 PLP AP Pap Pap TA TP a orsffo l/s, Ble fais o yl-2 1} Jo i/o IC [le Ti os Ile rp IL These nine matrices may be multiplied together to give one two by two matrix k'} (2 4 relating y' a' and y a at the planes QQ and PP respectively. If the plane PP By y Ps is moved to the vertex 0 and QQ to 0" then we have the matrix, 2. off2 Jf. of fi Pr of farypi o o 16} fo y o uf f-a 15} fo aff-2 2 10 i 1.5 which represents the effect on a ray in going from the plane through the first vertex 0 of the system through to the plane through the final vertex 0' of the system. It is useful in describing the general optical system to base our discussion on this matrix which we will designate as oo Obtaining the elements of the matrix may involve a lengthy series of matrix multiplications as in our example above, involving 7 matrices . However, all the necessary information about the optical system is included in this result~~22- ant two by two matrix and from it the cardinal points and planes of an optical systen are readily obtainable. From these cardinal points we may completely describe the image-forming characteristics of the system. In general then, we can assume the situation of Fig. 11. We know (or can compute) the matrix relating the rays going from the plane through 0 to the na] [\- 222] gy ‘Thematrix equation relating the emergent ray (y'a') at plane QQ a distance D' from 0' to the incident ray (y a) at the plane PP a distance D to the left of 0 4s y" . 1 opt ayy 1 oD y at o1 a, a, o 4 a Pe : or 00 Q YY] faut a PC, + PtHP") Pe a fy ayy + D ay aly @@ P vol Je - In Fig. 10 we had D' = 2 cm and D = 3 em. We are now in a position to discuss the first two cardinal points (and planes), the first and second focal points (and planes). The Focal Points and Planes If a beam of parallel rays enters an optical system the emergent beans will converge toward a point (or diverge avay from a point for a negative system). This point will lie in the second focal plane. If the rays are parallel to the optic axis of a system the point is the second focal point. The focal plane passes through the focal point and is perpendicular to the optic axis. This is illustrated in Fig. 12. We shall now determine the position of the second focal point, F', in terms of the elements of the matrix [A],,. To do this-23- OF aaNet= 2 - Figure 11 Figure 12,= 25 - let us take our second reference plane QQ to be the second focal plane in Fig. 12. Then D' = O'F', The first reference plane PP can be anywhere to the left of 0, We now note that every ray of the parallel beam passing through PP no matter what its y coordinate is, passes through the same point y' in the focal plane. That is, one of the linear equations relating y} a} to y,a must be of the form y! = (constant) a. Thus, the upper diagonal element of the matrix of equation (22) must be zero so that, DY = ort = = ay fay, (23) ‘The position of the second focal point, F', relative to the second vertex 0! is thus determined in terms of the matrix elements relating 0' to 0. A positive value of D’ indicates that F' is to the right of 0"; a negative value means F? is to the left of 0’. Rays emerging from a point in the first focal plane are illustrated in Fig. 13. A parallel beam emerges from the optical system. We can take the first reference plane PP through F and QQ the second reference plane anywhere to the right of 0'. Now we note that the rays through Q’ pass this plane at constant angle a although the incident rays all have different angles a at the plane PP, the focal plane. So one of the linear equations must state that a! is independent of a, (and depends only on y). Equation (22) gives this relation if the lower diagonal element is 0, that is, if D=O0Fe-aylay - 4)- 26 - Here D is positive if F is to the left of 0, negative if F is to the right of 0. Then we have from equation (22) that oeayy . 25) Thus the focal points may be easily found from (23) and (24) once the matrix [A],,1 is computed. ‘The Matrix Relating the Focal Planes It is extremely useful to choose the reference planes PP and QQ to be the first and second focal planes respectively. The matrix equation relating the emergent ray y', a', at the second focal plane to an incident ray y, a, at the first focal plane is given by Hebel F FF! F. Figure 13-27- From the discussion above, the matrix has a particularly simple form, since both diagonal elements are zero. Using (23) and (24) in (22), we obtain Ty r - daly of tf “|. ea] as) Te [a o| llr: L we The significance of the two non-zero elements of the matrix connecting the two focal planes is not evident at this point. However, as will be shown later (page 36) the elements are related to the first and second focal lengths, f and £* , defined as the distances from the first and second principal points to the first and second focal points respectively (see Fig. 17). The principal points are defined later as well. In che familiar case of the thin lens, the two principal points are coincident at the lens centre. For the moment we can anticipate the later results by designating the two non-zero matrix elements of equation (26) as 1 a --k (28) and lle, 3) a2 A very general relationship between f and f' may be easily shown by using the theorem on the product of determinants of matrices. First let us note that the dererminant of a transiation mrix, |? °| , te unsey end the deterainant of the refraction matrix separating two aedia of index n, and op s6ys 18 10 2 a a. Now referring to our example of Fig. 10, connecting the planes PP and QQ the determinant of the product matrix is given by: a 3 m8 ignoring all the translations as they involve multiplication by one. (In the- 28 - example we had n= n' = 1). A translation of the planes PP and QQ to the first and second focal planes respectively does not change the value of the determinant. Thus the determinant | 9 f= (30) 1 9 F : i, The ratio of the first and second focal lengths is seen to be equal to the ratio of the index on the left of 0 to the index on the right of 0'. Also f and f" must have the same sign. Obviously this result can be extended from the example of Fig. 10 to include any nunber of surfaces. Equation (30) illustrates the power of the matrix method,as the result is difficult to demonstrate in other ways for even a single thick lens and a general proof is not usually attempted for more complicated systems. It is also perhaps well to note that the theorem on matrices constitutes a useful check on the accuracy of multiplication in computing (A] 444. In the example of Fig. 10, the final product matrix (involving 7 matrices) should have a determinant equal to unity since we had n =n! = 1 xample A thick biconvex lens in air has a radius of curvature of 10 cm at each surface and the surfaces are 10 cm apart. Find the focal points and the focal lengths of the lens if the index of the glass of the lens is 1.50. Solution: ojji 10 1 0 LS 1 1-10) “1 jjo 1 [2/3 20/3 |-sre0 2/3 . 60 3 OF = 2/3(12)= 8 cm. O'F! = 2/3 (12)= 8 cm. [Align — we fle 12 m= £ since n=n'= 1. Fis 8cm to the left of 0, F! is 8 cm to the right of 0", Exercise 1 Take the lens of the example to be 1 cm and show that £! = £ = 10.17 cm, OF = 9.83 cm, O'F! = 9,83 cm.- 29 - Exercise 2 Take the lens to be a biconcave thick lens with radii of 10 cm , index 1.5, thickness 1 cm. Show that £! -9,.84 em» OF = -10.16 cm, O"F! = 10.16 em . Exercise 3 A thick lens has 2 convex surfaces, the first of radius 10 cm , the second of radius 20 cm. If the lens is in air and has an index = 1,5 and the thickness =1 cm ,show that £' = £ = 38.71 cm , OF=39.35 cm , O'F' = 39.42 cm. ercise 4 Consider the thick lens of the example above in which water (of index 4/3) is to the right of 0'. Show that f' = 240/11, f=180/11, 0'F" = 160/11, OF = 160/11. Exercise 5 Find the matrix [A},,, and the [matrix],,, for a thin lens (that is, the thickness of the lens is taken to be zero, with index n on the left, n’ on the vight and the index of the lens,y. Take the radii of curvature as R, 1 and Ry~30- Chapter 5 Image Formation lewton's Equations In Fig. 14 is illustrated the formation of an image point at y' of an object point at y by an optical system between 0 and 0'. The plane PP repre- sents the plane through the object point, a distance x to the left of 0; QQ is the plane through the image point, a distance x' to the right of 0'. ‘The focal points F and F' can be determined as discussed in the previous chapter and the matrix relating a ray passing from the first focal plane to the second is given by | 0 € FE? : ‘The planes PP and QQ can be related by using this matrix and two further translation matrices, Thus a ray y, a,at PP emerges as the ray y'y', at QQ such that fa x4] o£ 1 x] fy a o if fe o ou a or FFT Pr TP GD yy oe ve TF Y @ a -x a oLf r PQ Poo (32) Since all the rays emerging from y arrive at y’ regardless of the angle a, the upper right hand element of the matrix of (32) must be 0, in which case xxtefet (33) and yl = - (x'/£") y (independent of «) : Ga It is clear fron Fig. 14 that an object of height y is imaged as an image of height y'. The lateral magnification, m, is defined as the ratio y'/y which from equation (34) is given by-31- ma ytly = -x'/f! = f/x : (35) Equations (33) and (35) are Newton's equations for image formation and apply generally to any optical system. Example 1 An object 2 cm high 1s placed 40 cm to the left of the vertex 0 of the thick tens of the example on page 28. Find the image position and image size. Solution: Since F is 8 cm to the left of 0, x = 32 cm 32 x" = (12) (12) 4.5 em. Or the image is 8+ 4.5 12.5 cm to che right of 0’, The magnification, m, = -12/32 = - 3/8 and so the image is inverted and is 2 x 3/8 = 3/4 cm in size. In summary, we are now able to find corresponding image and object positions for any optical system by; Determining the product of the refraction and translation matrices between the first vertex 0 and the last vertex 0", the product matrix being designated as [A], o4- opricaL| SYSTE} Figure 1é= 32- 44, Obtaining from [A], ,, the position of the focal points relative to the vertices, (the distances OF and 0'F' from equations (23) and (24))as well as the focal lengths f and f' from equations (28) and (30). id Using Newton's equation (33) to compute the positions of the image and object relative to the focal points and equation (35) to compute the magnification. We shall use the following example to illustrate further these steps and later use the example to illustrate other cardinal points which are discussed in the following sections. Example 2 The thick lens of Fig. 15 has radii of curvature equal to 1.5 cm and 00! is 2cm. The index of the lens is 1.6, the index to the left of 0 is 1.00 and the index to the right of 0" is 1.30. an object is placed 17.167 em to the left of the vertex 0. Where is the image and what is the magnification? n= 1.00 at = 1.30 Figure 15Solution: qa) or ay (iit) -33- 25) 13 26 = 4,333 em (from equation (28) ) . TT gly B= 32 - 3.233 cm (£rom equation (20) )- 3 ter 3 o'rt = =§ = 2.167 cm (from equation (23) ). oF = 5 = 4.167 cm (from equation (24) ). (See Fig. 16.) (Also illustrated in the Figure are the principal points, H, H', which will be discussed in the next section) The object is placed 17.167 cm to the left of 0 so that x = 17.167 ~ 4.167, or x = 13 cm. ' fe" 3. 1 = FE Ay ~ cw (trom equation (33))+ the ‘ 10 59 image is at a distance > + 2.167 = 22 om to the right of 0". xt = as =- =H (from equation (35)) =i. =7 0 a £* “39 The Cardinal Points Tt is often convenient to discuss image formation in terms of other= 35 - positions than the focal points. These other points are the positive and negative principal points and the positive and negative nodal points. These, together with the focal points,are all defined as the cardinal points of a lens system and the planes perpendicular to the optic axis through these points are the cardinal planes. From the basic definitions of principal points and nodal points, it is a simple matter to find their positions using the matrices already described. ‘The positions of the principal points and nodal points are most readily specified in terms of their positions relative to the focal points F and F*. Let us assume that a cardinal point Z, say, is a distance z from F and the cardinal point Z' is a distance z' from F". Then the ray y', a’, at the plane through 2" goes through the plane at Z with coordinates y, a. This can be expressed as Zz. (6) We may now define the principal and nodal points and the definitions will impose conditions on the matrix of (36) which will give the positions 2 and 2‘ of the points in question relative to F and PY. Positive Principal Points The planes through the positive principal points, H and Hl", are the posit- ive principal planes and are defined as planes of unit positive magnification, te ye (ay a. that is,planes for which y' = y. From (36), “ Thus we must have 22'/f' = £ and -2'/f' = 1 for the positive principal points- 36 - wand HT, Then (37) (38) and 0 y £ a . Fl HH" H (39) The first and second focal lengths are generally defined as the distances FH and F'H' respectively, We see then from equations (37) and (38) that we are justified in having written equations (28) and (29) on page 27 where we introduced the symbols £ and f' into the matrix connecting the first and second focal planes. The principal point, H', 1s a distance £' to the left of F', the principal point Ha distance f to the right of F for a positive focal length system as illustrated in Fig. 17. Depending on the system the points may lie on either side of 0 or 0" and even may be "crossed," that is,H' may lie to the left of H. 1 ! I 1 | | ' Ft © t | | | ! | F # ! $ Komen le gs o—4 I oan 1 1 t Pigure 17-3- In Fig. 18 is illustrated a ray diagram showing the formation of an image using the properties of the principal planes and of the focal points. (Equat- ion (34), y' = =x" y/£", also follows from the diagram.) The object distance as measured from H is 8, the image as measured from H' is 8’. The ray y$a$ at the image is related to the ray Y, a, from the object by oe) s] fy {fo af fiael fo a} fa “is! - ff HH? 1-8! sétzs') + stf7 - = = y _4 att “fo = $F “Jo, (40) As stated before in deriving Newton's equations, the upper right hand element of the matrix must be zero if y! is an image of y, that is, if the rays through the image plane at height y' are to be independent of a, ‘Thus we have ft + ap which makes the upper right hand element zero and gives y' = my, where m is the magnification, or I Sh | x sle—t a ie | SS Figure 18= zis! vast ne Fs (42) Equation (40) becones 2 ° ¥ . £ a . (43) ample 3 Use (41) and (42) to compute the image positions of the object in example 2, page 32. Solution: From the results of example 2, (see Fig. 16) H is 3.333 cm to the right of F or (4.167 - 3.337) = .833 cm to the left of 0. H' is 4.333 cm to the left of F' or (4.333+2.167) = 2.167 cm to the left of 0'. The object distance S, = 17.167 - .833 = 16.334 cm (= “5 cm ) and L sgt @ ow wate Bow, : is ( 2 59 : Measured from 0' the image is ( 97 2.167) = im (to the right of 0') in agreement with Newton's equations of the previous example. - 42 Atso oo FL -a 3.333 -10 bet a= 3 ' 3 4.337 39 «38 before. By If the mediumhas the same index on either side of the optical system, f = £' and these equations take the very familiar forms, he ale 222 fo FE , (4a) and a (4s) For a thin lens in air, the situation is even simpler as the image and object distances s' and s can be measured from the same point, the centre of the lens, since for a thin lens the principal planes are approximately coincident at the lens centre. This and other special cases will be discussed in the next chapter.-39- Negative Principal Points The planes perpendicular to the optic axis through the negative principal points, H(-) and H'(-), are the negative principal planes and are defined as planes of unit negative magnification, that is, y* - It is left to the reader to show that z= F'H'(-) = £', (46) z= FH (-)=£, (47) and PF ~1 0 y elt faa [- B'CL FF a (-) . (48) BDH) Thus in the positive system illustrated in Fig. 19, an object placed at the distance £ to the left of F will give rise to an inverted image, the same size as the object, at a distance £" to the right of FY. It is left as a further exercise to the reader to show that the equation for image formation may be expressed in terms of the object distance, u from H (-) and the image distance, u', from H'(-) as = cag) and as (50) Figure 19= 40 - The Positive Nodal Points The positive nodal points, N and N', are two points lying on the optic axis which have the property that a ray incident on N emerges from N' in a direction parallel to the incident ray. Thus the angular magnification a'/a = 1 for these two points. To determine the position of the nodal points, we again use (36) but with the conditions now that for y'= y= 0, a =a, Equation (36) states that for y = 0, a' = (-2/f") a and therefore z= Mnf (st) Also (36) states that for y'=y=0,0= (2 + :) a which combined with equation (51) gives gta BIN' = : (52) Thus for a positive system the first nodal point is a distance £" to the right of F and the second nodal point is a distance f to the left of F'. The use of the nodal points in a ray diagram is illustrated in Fig. 20, The distances NN' and HH" are obviously equal. If £ = £', N and H are coincident, as are Nt and HY. Equations (51) and (52) together with (36) give the trans formation from one nodal plane to the other as y! f 0 y fF a’ 11 £ wnt (53) Again the two planes through N and N' are recognized to be object - image planes as there is a zero in the upper right hand corner of the matrix. ‘The linear magnification for an object at the first nodal plane of an image at the second nodal plane is seen to be m= y'/y = £/f" = n/n’.-41- Figure 20 Tt is left to the reader to show that the equations for image formation may be expressed in terms of the object distance v framN and the image distance v! from N' as fa (54) (59) and aoe e : (60) The ray diagram of Fig. 22 illustrates the use of the negative nodal points, The parallel rays through F and N(-) are brought to a focus in the second focal plane while the ray through F is rendered parallel to the optic axis. Exercise Compute the positions of all the cardinal points of the lens in the example on page 28. Compute the positions of all the cardinal points of the lens in exercise 4 on page 29. ee Ke | | F aofw +4 * Bigure 22~ 44 Chapter 6 Practical Optical Systems The methods developed in the preceding chapters may now be applied to a number of common optical systems. In each case we shall compute the matrix (Alo: from which all the cardinal points are obtainable. We shall mainly use equations (23), (24), (28), and (30) which give the positions of the focal points and the focal lengths. Then Newton's equations (33) and (35) may be used for image formation,or alternatively equations (41) and (42) which make use of the principal points given by equations (37) and (38). The Single Refracting Surface For a single spherical surface separating two media of indices n and n', the matrix [A],,1 is simply the refraction matrix, namely, 1 ° aR We can immediately recognize fle ak . (oD and pee (62) The distances ort = %R a gt nn . and oF = Since (37) gives F'H' = -f' and (38) gives FH = -f, the principal points H and H! are here coincident at the vertex 0 of the refracting surface. Example 1 As an example, consider exercise 1 on page 18 in which an object is placed 24 em to the left of a vertex 0 of a convex spherical surface separating media of index 1.00 and 1.50. The radius of curvature of the surface is 6 cm= 45 - (see Fig. 5). Solution: = Wea, f= 7, (18) = 12 cm OF’ = 18 cm x = %- = Bem et = GIO2D 21g en, 12 The image distance (from 0 to the image point) is 18+ 18 = 36 cm. In this example the image and object positions are in fact the negative principal points since m= -x'/£" = -18/18 = -1, The nodal point N is a distance f" to the right 2'R aR e R= 6am. O£ course N' is coincident of For ON=f'-£ with Ny as H' is coincident with H . This example is illustrated in Pig. 23. Figure Example 2 We can do example 1 on page 31 by repeated application of the above equations for a single refracting surface. We had a thick biconvex lens, index 1.5 in air, 10 om thick, with radii of 10 oq . An object was placed 40 cm in front of the first vertex 0.~ 46 = For the first refracting surface, p= GG) 2 39 on, +3 1 £ = G5) (30) = 20 em, and 20 2 60 4 Ge = 1 gives 8! = 60 cm. This gives an image 50 cm to the right of the second vertex 0' so that the image distance for refraction at the second surface is 8 = - 50 cms For the second surface, ee G)G10) 1). 2 em , f = (1.5) 20 = 30cm , and 20 4 wLgives ate 125en in agreement with our previous result. And the magnification, so) /-az.s o-- (2) (22) = = gas before Exercise 1 Find the first and second focal points of the lens of example 2on page 32 by repeated application of the equation for image formation by a single refracting surface. We might note that the plane surface is simply a special case in which R= © and the refraction matrix is 1 0 0 4 The focal lengths are equal to. Tt is easy to show that a) if s is the object distance measured from 0 and s' is the image distance also measured from 0. For example s' represents the "apparent depth" of an object a distance s under water (index 4/3) say, such that st = 3/4 8, Exercise 2 Derive equation (63).47 - The Thin Lens Possibly the most useful equations of geometrical optics are those for image formation by a thin lens. A thin lens of index y, is shown in Fig.24 , with media of indices n and n! to the left and right of 0 and 0" respectively. The thickness, t,of the lens is assumed to be small compared to the focal lengths of the lens such that we can set t = 0 in the translation matrix from 0 to 0". Thus 1 0 a) fal = oo! oo. or 1 0 ty oe fusat pew a oot PSR + BB, ia"R, ntRy at F ' F Radius = Radius = 'R, Figure 24= 48 - (assuming 0 and 0" approximately coincident). So, 2 = oF - - (64) Equation (64) reduces to the well-known lens-makers' equation when n' = n 1.00, nanely, ala 1a $erd-ow(z - 2 : (65) 1 Ry Exercise 3 Show that the points H and H' coincide at the assumed approximate coincidence ©£ 0 and 0" which we take to be the lens centre of the thin lens. Exercise 4 Find the positions of the nodal points of the thin lens. Newton's equations apply here using (64) for f and £'. It is also convenient to use (41) since sand s' are measured from the centre of the thin lens, that is, sie ne (= waren) The matrix representing a thin lens has the particularly simple form (66) Question A zero in the upper right hand corner of the matrix means that the matrix relates an image plane and an object plane. What does the zero mean in Equation (66)?- 49 = Combination of Two Thin Lenses The combination of two thin lenses in air forms the basic arrangement for the most common optical systems such as the microscope and astronomical refracting telescope. Also, most oculars are conbinations of two lenses as corrections for some of the defects of images can be reduced in this way. The defects are classed in terms of five monochromatic aberrations arising as correction terms to the first order theory, plus two so-called chromatic aberrations arising from the dependence of the index of refraction on frequency. The combination of two thin lenses in air, a distance d apart,is shown in Fig. 25. We may take the vertices of the system as the lens centres 0, and 0, The matrix is then, using (66) with £, =f}, and £, = £5, 1 0 loa 0 A000, 1 1°2 l oO o1 1 fy Figure 25= 50 - Therefore if £ (= £') is the focal length of the combination, equation (28) gives, All gal en f"#, * £, : The focal points are F and F' such that oO) F=£ (l- 4/t,) (68) and fede). (69) using (23) and (24). One simple special cese of interest is that in which two thin lenses are cemented together to form a doublet. In this case d= 0 and $=+ +2. 1 fy Such a doublet is usually constructed to reduce longitudinal chromatic aberration. the lenses are made of different glasses which have different dispersive powers (i e. the indices of the glasses depend differently on the frequency of the light) and so the combination may be designed so that the focal length £ will have a minimum dependence on the frequency of the light (Jenkins and White, p. 157). It is also possible to reduce the dependence of the focal length, f, on the frequency when using two lenses of the same index of refraction by choosing a separation d=(f, + £,)/2. This may be readily shown using (67) and (65) and solving for a minimum dependence of f on, the index of the glass of the lenses (Jenkins and White, p. 163). ‘The Telescope ‘The telescope in its most basic form is a special case of two thin lenses in air. In this form it is a simple astronomical refracting telescope. Basically, a telescope is used to increase the angle that a distant object subtends at the eye. If the eye is relaxed for distant viewing, the-sl- telescope simply produces an angular magnification in which an incident (approximately) parallel beam from a point in a distant source, making an angle a say with respect to the optic axis, emerges as a parallel beam making a larger angle a' with respect to the axis. When the object is viewed with the instrument, the image formed on the retina of the eye will be larger by an amount proportional to a'/a, the angular magnification, M. This telescopic system is characterized by a zero in the lower left hand corner of the matrix, [a], 4,» which we obtain if d= £, +f). Then, from [A], | above, 1°2 e192 : £2 y (er + £9) (70) or M=a' /a=- f/f. Thus the angular magnification is 2s the negative ratio of the focal length of the first lens, the objective lens, to the focal lengths of the second lens, the eyepiece lens. As illustrated in Fig. 26 the image as viewed through the telescope is inverted, and the image formed by the objective lens is in the second focal plane of that lens which is also the first focal plane of the eyepiece lens. —-J4-—- Figure 26For a terrestrial telescope one may insert an erector lens in between the objective and eyepiece lenses such that the image formed by the objective acts as an object for the erector lens which in turn forms an inverted image at the first focal point of the eyepiece lens. ‘The matrix of this system of three lenses will have a zero in the lover left hand corner and the angular magnification is positive. Exercise 5 Find the matrix of the system of the three lenses of Fig. 27 and show that they form a telescopic system with positive angular magnification. Solve for the case in which f = 20 cm =2cm and f)= 5m. (answer M = + 4) Exercise 6 A Galilean telescope has an objective lens with £ = 20 cm and the eyepiece lens with f= -5 cm. The lenses are separated by 15 cm. Calculate the matrix for this system and show that it is a telescopic system with M= + 4, coheel y Y ra e— 2fpsle tho J, Figure 27-53- The Ocular In most telescopes it is common practice to have a system of at least two lenses called an ocular to perform the function of the eyepiece lens. The first lens or field lens improves the field of view and in combination with the second lens or eye lens can be chosen to reduce some of the aberrations inherent in a single lens. The most common oculars are the Ramsden and Huygen types. Exercise 7 A Ramsden eyepiece consists of two equal focal length lenses, f,. If the separation is equal to £,, chromatic aberration is minimized. In practise the lenses are usually moved closer together (Jenkins and White, p. 182). Calculate the position of the focal points and the focal lengths of the ecular in the case in which d = 3/4 £,. ani fiers The ocular discussed above as well as just a single thin lens can be used as a magnifying glass. When an object is examined an image is formed at the retina in the eye and this retinal image may be increased in size using a magnifier. For continued viewing, it is usual to assume that an object 25 cm in front of the eye can be comfortably viewed. For this case then, an object of height y subtends an angle 6, at the eye with 6, = y/25, as in Fig. 28. Now 4£ the object is placed at the focal point of an ocular, the energent rays from the ocular may be focused by a relaxed eye as also shown in Fig. 28. The rays from y now make an angle of 8, = y/f where £ is the focal length of the ocular, Thus the change in the size of the retinal image is given approximate- doy ft le Lacie, “o ¥/25 The ocular may also be used as in Fig. 29 with the object just inside the focal point of the ecular for an eye accommodated to the 25 cm viewing distance. In this case the magnifying power (which is usually quoted on the~ 54 = eyepiece) is (1+ 25/£). This is an approximate formula which assumes that the image distance from the second principal point is 25 cm and then % Wl, 5 west -28 = Be 95 ¢ % Yas § (since s' = -25 om). 1 Exercise 8 Calculate the magnifying power for the Ramsden eyepiece of the above exercise 7 for both a relaxed eye and an accommodated eye if f, = 2 cm. x y af F H} Ht Pe nas 2 SSN | NAN | SS NON | NOS | NTS ! y N i ne = Figure 29- 55 ~ Chapter 7 Spherical Mirrors Convex and concave spherical reflecting surfaces have image~forming qualities which can be discovered from the laws of reflection stated in chapter 1. In what follows we shall apply the matrix method to the case of reflection, again limiting consideration to paraxial rays. The difficulty that often arises in discussing image formation by mirrors is that of applying the sign convention used for refraction. In the case of refraction the rays continue in the same (positive x) direction after refraction but in reflection the rays are of course reversed. To avoid the difficulties,one may define a different convention for reflection or specify the positive image direction in terms of the on-going emergent rays. Or, in terms of the mathematical formulation, one can regard reflection as a special case of refraction in which Snell's law, np = n'g",(for small angles), becomes the law of reflection, ¢ =~ ¢',for n= - a’. In the matrix formulation it is perhaps easier to maintain all our Previous sign conventions for refraction and treat the reflected rays as having a slope angle equal to ~ a’. This in effect allows us to treat the rays after reflection as if they were proceeding on in the same direction as the incoming rays. Of course the final calculated quantities such as a positive image distance or a positive focal distance,for example,will actually Lie to the left of the reference point in constructing a ray diagram. The actual path of a ray after reflection from a convex mirror is shown in Fig. 30 and the corresponding diagram for deriving the reflection matrix is shown in Fig. 31. ‘The transformation equations for reflection are easily derived from Fig. 31. From the figure,- 56 - Figure 30 Figure 31-57- and g'=e4+6- +a Ro since 5 = ¥/g = % /R, The transformation equations are then, where | The Single Reflecting Surface For a single reflecting surface the [A],,, matrix refers simply to the oy or is the reflection 1 mine matrix. vertex of the reflecting surface and from (28) with [A], equal to the reflection matrix, fl =-R/ 2 (my and from (30) £' = f as the matrix determinant is unity. Of course Newton's equations for image formation apply as always and may be written here as exter? / 4 and R/ 2x = 2x" / RK Alternatively, ££ 21 Secomes tyi. a sits >and The distances x and x', sand s', and R follow the same convention as before for refraction. The final answer however must then be "reflected" back to its true position. Let us illustrate this with the following example: Example 1 A concave mirror has a radius of 12 cm and an object 2 cm high is placed~ 58 = 24 cm to the left of the vertex 0 of the mirror. Where is the image formed and how large is it? Solution: (1) £=-R/2= cm “CD og OF is 6 cm to the left of 0 OF! is 6 ch to the right of 0 . Since x= 18 cm, x" (18) = 36 x'= 2m to the right of Ft or 8m to the right of 0 (I"(calc) in Fig. 32) . Now the actual rays pass through the point 8 cm to the left of 0, the point I' in Fig. 32. The magnification m= ~ f/x = - 6 / 18 and therefore the image is inverted and 2/3 cm in size. @ We may also use 14a. teh ef Z,+3 = “7g which gives 1.2. 2 .ii di s'" =(2)~ 2% 6 24 . Then s! = 8 cm to the right of the matrix plane at 0 (I'(cale) in Fig. 32) or the real image is 8 cm to the left of 0, (I' in Fig. 32). XI" (ale) \ Figure 32- 59 - A Thick Mirror A thick lens with the second surface covered with a reflecting coating as in Fig. 33 can be treated by the matrix method. The rays on reflection are considered by the reflection transformation above to be proceeding in the positive x direction so that the optical system to the left of the reflecting surface at 0’ must be "reflected" at 0'as well, as shown in Fig. 33. The final distances in the problem are referred to 0", that is,a positive distance to the right of 0” is actually the same distance to the left of 0. xample 2 A biconvex lens 2 cm thick with a radius of 6 em is silvered on the second surface. The index of refraction of the lens material is 1.5. Where are the focal planes of this thick mirror? Figure 33- 60 - Solution: 1 On Afr FA Am go (Al jou = oo" “} 2 3 2 a 3 12 2 ooL 3 4 oy 18 3 3 1] a 3] 7) aa 5 a oF Thus £' 81/44 cm =f. OF = (81 / 44) (5 / 27) = 15 / 44 cm using equations (23) and (24) (Page 25). O"F = 15 / 44 em, The ray diagram is shown in Fig. 33. Example 3. As a further example of the use of the focal planes, suppose an object 2 cn high is placed 15 / 22 cm in front of the vertex 0. lhere is the image and how large is ic? Solution: , as of : or 9.94 +15 = 10,28 cm to the right of 0", which is actually 10,28 em to the left of 0. 3, : ma 162 ny Since m= gilt =~ 5.4, che image ts M2 om in size and inverted “ (or m= -x'/f" =~ 9.94/1.835 = - 5.4). Exercise Solve the above example to find the focal positions using repeatedly the equations for reflection and refraction at a single surface.-61- Astronomical Reflecting Telescope One simple form of the astronomical reflecting telescope is shown in Fig. 34. A parallel beam of light reflected from the spherical concave mirror is brought to the second focal point, Ft,of this mirror after further reflection by a small plane mirror. The focal point Fy is also the first focal point F_ of the eyepiece lens. The plane mirror is sufficiently small so that it interrupts only a small amount of the light falling on the concave mirror. Tn another conmon arrangement, the spherical mirror has a relatively small hole in its centre and a small plane (or convex) mirror redirects the Light back to a focus at the centre of the large mirror where the focal image is examined with the eyepiece. Figure 34~62- ‘The matrix method may be used to discuss this telescope as indicated in Fig. 35. Then, if fy = fy =~ R/2 equals the focal length of the mirror and ff] equals the focal length of the lens 2 7 A ep. a [gor 7 1 1 (alogr = “fy 0 L . The telescopic system is again indicated by the zero in the lover left cr, hand corner of the matrix and the angular magnification, M, = a'/a = t=, . FT wae, In practice, the large reflecting telescopes are usually used simply as cameras with a photographic plate placed in the focal plane of the large reflector, The surface of the reflector is made parabolic in shape to avoid spherical aberrations, Of course a reflector ts free from chromatic aberrations as the laws of reflection are the same for all frequencies of light. The mirror may be made large in diameter in order to attain a high angular resolving power and to increase the light flux per unit area (the illuminance) at the photographic plate. “7 Figure 35-63- APPENDIX Generalization of Matrix Optics for Arbitrary Ray Directions. In all our earlier considerations we have tended to vse rays specified by the matrix [ ¥ ] which are coplanar with the optic axis, perhaps even without causing concern to the student. But of course all (real) rays are not necessarily coplanar with the optic axis and their description at any plane perpendicular to the optic axis is in terms of four variables, namely y, a, z, 3 where z and & refer to distance and angle from the optic axis in a plane perpendicular to the usual measurement of y and a. A diagram will be helpful in explaining the convention and will demonstrate the translation matrix in our new system. Referring to Fig. 36, the planes P,P,P,P, and Q,0,0,, represent two parallel planes perpendicular to the optic axis (x-axis) sepa~ rated by a distance D, The position y, z and y', z' of the ray EF and the projections of this ray on the xy and xz planes make angles « and 2 with the x-axis. We note immediately that yl=y+ Do tea tDs~ 64 -- 65 - for which the matrix representation is Foy 1p 00 at o 100 « 2 o old 1 Lost oo 01 18 a ~ m > +P Tt is obvious chat the large matrix consists of two identical nested translation matrices, the remaining elements being zero, implying that there is no coupling between the variables y, a and 2, 6. Working out the refraction matrix for a spherical surface using paraxial rays is a bit more difficult. Say we have a ray defined by y, a, z, 8 in a medium of index n entering a surface with radius of curvature R and index n'. The angle that the surface normal makes with the optic axis is given in terms of its projections on the xy and xz planes, namely «, and 6, where a, = 3 22 and ee ‘The angle that the incoming ray makes with the surface normal can also be broken into two components, namely sata, L BHR. ‘These components can be taken in the coordinate system referred to N because the angle between N and the optic axis is small.~ 66 - Referring to Fig. 37, 8 and 8" represent the angles of incidence and refraction with respect to the surface normal N. We av at a2 +e2 =e? , have then that a2 + @2 = @ Now n' 6! = m8 by Snell's law for paraxial rays and the incident and refracted rays must be coplanar with the surface normal as well. Hence the exit angles with respect to the surface normal a} and 8} satisfy equations aC at? eet?) = ne Car + ee Cay? + 812) (a +62) and aL evi - 7 giving a solution By = @tap Boe ta = OtBy These angles of course refer to the surface normal, so to take then with respect to the optic axis, we must subtract a, and 8, respectively = @ta)-a Ey 8 st=e'-8, = S (+a) -8, . 2,~67- Figure 37.~ 68 - We also have = -y' = y and so the matrix representation for a spherical refracting surface is Poa oO oO o fy } | zt oO 0 1 { poz | st 0 oO pal | 8 . Fortunately this again represents two identical nested refraction matrices and again there is no coupling between the variables. The student may readily show that the reflection matrix follows the sane pattern. We see then that any ray may be specified by its components y, a and z, 8 but each of these separately undergoes identical trans~ formation matrices with no coupling. Thus our earlier treatment is in fact a general one and we may safely regard our two by two matrices as a complete specification of the optical system. Of course if we used large angles and Snell's law in its correct form, we would get coupling between the variables and the matrix would have a more complex form, if indeed the problem could be linearized at all.KK— Optical system —+>| ge , AY NO) HO NO HO TF a + + x f kK f x Ku ul v > Ot Iw . w