rn NCE PVN a auc} 15 Circular motion 15.1 Horizontal circles 15.2 The 3-dimensional case 15.3 Vertical circles End-of-chapter review exercise 15 16 Hooke’s law 16.1 Hooke’s law 16.2 Elastic potential energy 16.3 The work-energy principle End-of-chapter review exercise 16 17 Linear motion under a variable force 17.1 Acceleration with respect to time 17.2 Acceleration with respect to displacement End-of-chapter review exercise 17 18 Momentum 18.1 Impulse and the conservation of momentum 18.2 Oblique collisions and other examples End-of-chapter review exercise 18, Cross-topic review exercise 3 Further Pure Mathematics 2 19 Hyperbolic functions 19.1 Exponential forms of hyperbolic functions 19.2 Hyperbolic identities 19.3 Inverse hyperbolic functions 19.4 Logarithmic form for inverse hyperbolic fun End-of-chapter review exercise 19 20 Mat 20.1 Eigenvalues and eigenvectors es 2 20.2 Matrix algebra 20,3 Diagonalisation 204 Systems of equations End-of-chapter review exercise 20 397 398 403 409 410 4 a7 429 430 431 432 438 440 442 446 447 448 451 461 465 an21 Differentiation 472 21.1 Implicit functions 473 21.2 Parametric equations 478 21.3 Hyperbolic and inverse functions 481 21.4 Maclaurin series 488 End-of-chapter review exercise 21 496 22 Integration 497 22.1 Integration techniques 498 22.2 Reduction formulae 503 22.3 Arc length and surface areas 509 22.4 Limits of areas 515 End-of-chapter review exercise 22 523 23 Complex numbers 524 23.1 de Moivre’s theorem 525 23.2 Powers of sine and cosine 529 23.3 The roots of unity 532 go 23.4 Complex summations 5337 End-of-chapter review exercise 23 543 24 Differential equations 544 24.1 First order differential equations 545 24.2 Second order differential equations: The homogencous case 550 24.3 Second order differential equations: The inhomogeneous case 557 24,4 Substitution methods for differential equations 565 End-of-chapter review exercise 24 575 Cross-topic review exercise 4 576 Further Pure Mathematics 1 practice exam-style paper 578 Further Probability & Statistics practice exam-style paper 579 Further Mechanics practice exam-style paper 581 Further Pure Mathematics 2 practice exam-style paper 582Hyperbolic functions In this chapter you will learn how to: c hyperbolic functions to the exponential function 3 of the hyperbolic functions use identities involving hyperbolic functions use the definitions of the inverse hyperbolic functions and use the logarithmic forms.ee ASU URES ea uC Aa aC eo sag ‘Where it comes from ‘What you should be able to do Check your skills AS & A Level Mathematics Use the trigonometric identities. | 1 Prove Pure Mathematics 2 & 3, Chapter3 cosa eae Gnd a8 GSEA Lia Mathanaies | Mnipulsc equatinsinohing [2 Sobeie*— Pore Mathomais 2&3, | he natral gat and Chapter 2 | exponential functions. What are hyperbolic functions? In the Pure Mathematics 2 & 3 Coursebook, Chapter 4, you learned uses for trigonometric functions other than solving problems based on a triangle. In fact, trigonometric functions are also called circular functions, as they all have geometric meaning and derivations from the equation of the unit circle, There are similar relationships between the structure of circles, hyperbolas, parabolas and ellipses, These shapes are all called conic sections, because they are shapes produced when we slice a cone. Horizontal slices produce circles; oblique slices produce ellipses; slices parallel to the slope of the cone produce paraboles; steeper slices than this produce hyperbolas. cece llipse parabola hypesbola ic functions are a class of functions that are derived from the unit hyperbola, ‘They are a significant tool in an engineer's tootkit. They also allow us to find highly elegant solutions to differemtial equations. 19.1 Exponential forms of hyperbolic functions In the introduction, we discussed the relationships between hyperbolic functions and trigonometric functions, Hyperbolies link to apparently vastly different function types, but as you read this chapter and do subsequent work on complex numbers, you will start to see the connections. ‘We shall start by relating hyperbolic functions to exponentials, It is not clear from the ideas and development of hyperbolics whether the exponential definitions came first or the relationships with trigonometry came first. We shall use the exponential forms to develop the idea. Let us begin with the hyperbolic sine function: inh x cos(d +B) sin Beos B 6CC td aay ‘The first challenge is the pronunciation of the hyperbolic functions. The hyperbolic sine is, pronounced ‘shine’ and its exponential definition is: sinh. Based on this, we can define the domain, find the range and draw the function. It is helpful to consider this graph as the average of y=e* and From this, we can see that the function is a one-one mapping. The domain is xR. The range is fx) eR. It is also an odd function. gi » The hyperbolic cosine is pronounced ‘cosh’ and its exponential definition is: cosh cosha = tS We consider this graph as the average of y=eF and y =e", From this, we can see that the function is nor a one~one mapping, However, it can be evaluated with both positive and negative values for x. ‘The domain is x € R. The range is f(x) > 1. It is an even function. For the function to be one-one, we need to restrict the domain to x > 0. This will be very important later.it sare Tena EI > 4 > : ryyes Et A Based on the exponential definitions of sinh x and cosh. find: ‘The hyperbolic cosine is a very special ccurve and has many occurrences in nature, It's called a catenary, and isthe shape of & hanging chain, 4 a £ (int) and 4 (cosh ax en a" The exponential forms of sinh x and cosh x are shown in Key point 19.1 P) key POINT 19.1 ‘The exponential forms of sinh and cosh: sine == XER, {eR een costy= 242", eR, f9)21 From this, we can deduce that e* = cosh x + sinh x We can now derive further hyperbolic functions “anch’(prelerred) cones ‘than’ (a Tong thhhhh as in ‘thank you’) f°" *~ er ee tank’ Domains x eR Range: 1 < fx) <1 ‘Odd funetion TUE aa Ure Cr Ueeta carta ec eter Domain: x €R sech “seteh" sech: Range: 0< fl) <1 Even function Ero ‘The graph of y =sech*y isa very important curve. Iti used to model tsunamis and to model wavelets in wave-particle duality. We cal this type of wave a soliton, It isa very special type of ‘wave that does not lose energy as it travels Its velocity is proportional to the amplitude of the ‘wave. This is why tsunamis have such a destructive force. You might study more about solitons at university evel -cosech x “cosetch” cosech x 2 = F Domainex4 0 4 3 i Range: fx) #0 m7 ee a4 ‘Osa function oth x ‘cath cot Damar 0 Toho ws remember the recprocl fictions of sechycosechxand cath, weean apy thethn terete Range: f(x) > 1,() <-1 swith trigonometric reiprocal functions: 1 sech x= cosh (Odd function peers fl sinbx 1 anh x ccotheit a caren OS 7 ea ua Cre a LL uC Enos a Sketch the graph of y= [Beosh(x +1) = 4 Answer Given that f(x) = cosh x Consider each transformation in turn. vette 1) = 341) 3x +1)—4 fx + 1) 4 This is the graph of [Beosh(x-+ 1) ~ 4).Ts CHOY erate ae ‘These functions can all be evaluated using a calculator, in the same way as for trigonometric functions. However, generally you will be required to write in exact form, unless told otherwise. We ean use the previous definitions to solve equations involving hyperbolic forms. roceiscnatstrd Leaving your answer in exact form, solve 4coshx ~ 3sinkx Answer Acoshx ~ 3sinhx = 4 32 4e 44343 = Be* -Be47=0 (e-7Ye"-1 Leading to two solutions: x= In7, x= in 1 =0 Iis possible to approach this problem using a similar technique as that for trigonometric equations using hyperbolic forms. However, this rarely gives such elegant solutions. We need to understand both inverse hyperbolic functions and their logarithmic forms Ea 3 Solve Stank? ~ dtanhx =4, Answer 3tanh?y—4tanhx—4=0 art ara. 37? 6r+27-4 30(T~2)+2(T = 2) (T-)GT+2) e Orr We are left then to solve: tanh = cm-(3)ene een Oe eta cure er ernst Write all solutions in exact form unless otherwise stated. 1. Solve 19sinhx + 16coshx =8. 2 Solve 29coshx = 11 sinhx +27, 3. Solve I7sinhx + 16cosh.x 4 Solve 7= I7tanhx + 28sech-. 5 Solve cosechx ~ 2eothx=2 6 Sketch the graph of p= [Beoshx ~ 5}, 7. Sketch the graph of y ~ cosech(x~ 2) 8 Sketch the graph of y 1 deoth(x +1 + jeoth(x + D) @_ 9 Solve 4cosh.x-+ sinh.x =8. Write your answers in the form b In a, where a and b are integers. @ _ 10 Soive 10coshx+ 2sinhx = 11, giving your answers in the form In a, where a isa rational number. ne Btanh?x = 1 @_ 1 Given that seh! 19.2 Hyperbolic identities If we consider that hyperbolic functions are similar (o trigonometric functions, itis reasonable to consider that hyperbolic identities exist in the same way. We can use the exponential forms to prove that they are true. ) soe dash When forming a proof, it is very important tocither start at the leftchand side and show itis the same as the right-hand side, or start atthe right-hand side to show itis the same as the left-hand side. Starting from both ends until you get 1 common expression will not make a formal proof. You can use this to help you see how 10 formulate your proof, Dut the final written answer must be in the ‘correct order, Make sare that you state every algebraic step: tell the whole story! cients Prove that cosh?x ~ sinh? = Answer ccosh?x ~ sinh?.x First, convert to exponential form and ‘consider the left-hand side. Expand and take outa factor of ‘This is the right-hand side. ‘Therefore, the left-hand side becomes the right-hand side. Therefore, cosh? — sinh?Chapter 19: Hyperbolic functions Eicon Beniats! Prove that sinh(4 + B) = sinh Acosh B + sinh Bcosh A Answer sinh Avosh B + sinh Bosh A Consider the right-hand side and convert it to exponential form. Expand and take outa factor of 4. ui, Use efePa et and ele? sinh(4 + B) This proof works from right-hand side to g Tefichand side, 8 required, You should have noticed that these two proofs are similar to trigonometric identities. It is possible to use Osborne's rule, as shown in Key point 19.2, to move from a trigonometric identity 10a hyperbolic identity. Nevertheless, this does not constitute a proof! Osborne's rule ‘To move from a trigonometric identity to 2 hyperbolic identity: change a cos to a cosh and change the sign of a produet of sinks. We can use Osborne's rule when we are required to use a hyperbolic identity without needing its proof. Praca end Using Osborne's rule, state the angle addition formula for tanh(A + B). Answer Start by considering the trigonometric identity: We consider replacing tan with tanh, tan + tanB tana +8 B 4+ eT tan Aten tanh + tanh tanh(4 + B)= sinh is the numerator of tanh. Applying Osborne’s rule, we notice that tanhdtanhB will require a sign change singe it includes a product of sinks, + tanh tanh B as required.Cambridge International AS & A Level Further Mathematics: Further Pure Mathematics 2 Brose © 1 Prove tat cosi*x cosech?x + 1 ccoth?x — 2. Prove that cosh3.x=4cosh?.x ~ 3cosh x 5 (tanh. ~ oth). © 2 Provethat coth2x ~coth © 4 Prove the following hyperbolic identities. a cosh( — B) = cosh Acosh B ~ sinh Asinh B son (458) 19.3 Inverse hyperbol {As with trigonometric functions, the hyperbolic inverse Functions must exist and be well defined. As with any function, the inverse function will be a reflection in the line y= x for the one-one aspect of the function (otherwise the inverse will not be well defined). You may have noticed by now that the notation for inverse trigonometric functions is a little ambiguous and it can be easily confused with reciprocal functions. There is another commonly used notation for them: b sinh3x = 3sinhx + 4sinbx sinh 4 ~ sinh B. functions eos"! x = arcoos x tan”? x= arctan x Pes ‘The origin of the prefix are-comes from the unit circle and the fact thatthe ar length is equal to the angle (J=r8, where r= 1), Finding the angle 9s then equivalent to finding the are Tength, So finding the angle, which requires inverting the sine Function, is equivalent 10 finding the arsine ‘There isa similar idea with hyperbolies, although beyond the scope of this course. Finding the hyperbolic angle is equivalent to finding twice the area of the hyperbolic sector of x? ~ 9? the unit hyperbotaChapter 19: Hyperbolic functions Tei The inverse hyperbolic funetions are ery re Graph erent St) sinhx sinh“! x, | workeD exams 19.7 | Ff Solve the equation sinh?x—Ssinh x+ 0. Give your answers to 3 significant figures, Answer sink? ~ Ssinhx +4 First, notice that this is a quadratic equation in terms of | (sinha 4)(sinx— 1) = sinh x, and it will fzctorise, When sinh x = 4,= sinh” 4: Make sure solutions are written to the required level of x=2.09 accuracy. When sinhx = 1x = sinh“! 0.881 Iis possible to write the answer exactly, using natural logarithmseee iC SEI OCe Tei a Cura uC LE auc rd 19.4 Logarithmic form for inverse hyperbolic functions If hyperbolic functions can be described in terms of exponential functions, then itis reasonable to assume that inverse hyperbolic functions can be described using the natural logarithm function. This is indeed the case and is shown in Key point 19.4 Eon EXAMPLE 19. Find the logarithmic form of sinh“! x. Answer Rewrite x in terms of sinh. Express in exponential form. Rearrange to make y the subject. To do that here, complete the square. ys Ine + V4 1) ‘Take logs to make y the subject Inge + ve¥ 7). Since we can only take logs of positive numbers, take only the positive root here, Consider why we do this rather than take the modulus. Therefore, sinh“ x no (ee Why do we consider ‘The logarithmic forms of the inverse hyperbolic functions are ‘only the positive root sink! In(x + Vx? +1) | here? L cosh"hx, In(x + Vx? - 1) tank tn(}=) oth ech"! cosech“! You need to be able to derive the logarithmic form for inverse hyperbolics. ‘You ate now equipped to give your answers in exact form. In the next example, we will solve the equation we had in Worked example 19.7 but this time we will give our answers in exact form,ap CH Ne aaa ey reciente: Solve the equation sinh? ~ Ssinhx +4 =0. Give your answers in exact logarithmic form. Answer sinh.x + (sinhx — 4)(sinhx - 1) =0 When sinhx = 4, x = sink" sinb=!4=n(4-+ V+ 1) x= In(4-4 VI) When sinh x sink] =In(1 + VI24 1) x= la( 4 ire isnuatstry sint"(AV1 + B? + BVT4 A). nh“! | Prove that sinh“! + sinh! Answer 1+ (4Vi4B? + BVI + ae =1424°B* + 24BVA + VBP +1 + A? + BP A°B + DABV A? + VB 41 + (4? + 1(B? + 1) [HAM 1+ B14 248 (484 VAS IVBR SP sinh"(AV1+ B+ BVI A) In AVi + B+ BVI + fae a + Vis avis B+ avi+ A) in(AVi + B+ BVI + A+ AB + VA =in((4+ V4 AB + V14 BY) =in(4+V1443)4+ In(B+ Vi +B?) = sinh! + sinh“! B as required, ‘There are some very interesting relationships between the inverse hyperbolic functions. We do not need to know these but they will help with some of the hyperbolic and trigonometric integration later. For example: sinh(cosh?x) = VF —1 03SoU Ue Sau eS eer aun cee u a LUC press 1. Find the exact value off a sinb?3 b tanh! © cosh! ) tant 2. Solve, giving your answers as logarithms, a 3cosh?x = —10sinhx b tanb?x + Ssechx— 5 ¢ sinh2x= coshx 3. Prove that cosh™x 4 Prove that tanb-!x Prove that sech~! x = In (wee 6 Prove that tanh! 4 + tanh"! B e000 0 0 7 a Prove that 4cosh'x ~ 3cosh x= cosh 3x, b Hence, solve cosh 3x= Scoshx cians su Given that sinhy = x, show that: a painget Vi+x) By differentiating this result, show that: wy p dx Answer ‘Use the exponential definition. Complete the square, poin(x+ VE +1) Rearrange and take the positive root,Ean) ions Therefore, | qu weet (14x9 Therefore, Simply. | | oye REET dy du “dx ay 1 Viee+ x > | = + xe Use watt xf av ye(4eh View 2 as required. | Vite | ere aT For inverse functions: © aresinhx = sinh“! x=In(x + V4 1) © arccoshx = cosh“! y= In(x-+ VP = 1) oth nts = ia *2)Ptseuranaas) a eT Sure Vea CCE Ucar aCe LUC Solve Seoshx ~ cosh 2x= 3, Leave your answer in exact logarithmic form. Starting from the definitions of sinh x and cosh x in terms of exponentials, prove that: a b 1+ 2sinh?x = cosh2x 2eosh2x + sinhx = 5 Show, by using exponential form, that the curve with equation y= cosh2x + sinh x has exactly one stationary point. Determine, in exact logarithmic form, the x-coordinate of the turning point,In this chapter you will learn how to: m formulate and ems of nultaneous equations with three unknowns utions of matrices geometrically to lines and planeseC Arn OS ee ai CCE ee Raa CL ured eee eor saad ‘Where it comes from ‘What you should be ableto do | Check your skills Chapter 4 Find determinants of matrices | 1 Find the determinant of each of the upto 3x3. | following matrices: 23) a |s 1 124 bB 2 0 la -1 -1 Chapter 4 Multiply matrices. 2. Work out the following calculations: a (2\(04 3 6)\-2 1 1-23\/023 bp (2 o1)(-121 4 33/L004 3 "Write down the normal of each of the following planes: Chapter 6 Find the normal of a plane. a Bxt2y-255 b -2x+52=6 c 3y=-ll What else can we do with matrices? In this chapter we shall look at a special type of vector called an eigenvector. | Eigenvectors are widely used. Applications include vibration models for bridge and building design and the page-ranking algorithm that is used to prioritise the results from search engines. We shall also develop our knowledge of matrices that we studied in Chapter 4. 20.1 Eigenvalues and eigenvectors 12 30, sos (8). (-3) an (2) snes as (3) (3 an (2) seer magnitude of the third vector increased, but its direction remained the same. ‘To begin, let us consider the matrix A= Lf we multiply this matrix by the So why do some vectors nor change direction? / ‘To answer this question, first consider the statement Ax = Ax. This says that a matrix A applied to a non-zero vector x results in a vector 4x. Since 2 is a scalar quantity, the direction of the original vector is unchanged,oem Varad ‘Now multiply both sides ofthe equation by the identity matrix to get IAx = 12x. Then Ax = lx, 100 4 0 020.0 which can be written as (A—ADx=0.Notethathere 21= | 0 0 i 0 OOH eee If (A — Al)? were to exist, then x would always be the zero vector, which is not the case. This ‘means that the inverse cannot exist. The only way the inverse cannot exist is if det(A — 21) We shall use this result to help determine the values of 2. This will enable us to determine 2 Lookin dc resis cxanls tine ih A=(! 2) ie We first find det(A — a1) = # ~ 2~6=0, This equation is known as the characteristic equation, and the 4 values which satisfy this equation are known as eigenvalues. The word eigen is German for ‘selP. = 0,and from here -2(1 - 4) - 6=0, or ‘Solving the equation gives 4 = —2, 3. We can use these values in the original statement: ren ane (1 2)(°)=a(2) oni ten writen a 5, x. This result implies For the case when 2 = a both equations give y= 2 thavany vetor ofthe form (_3) willnot change direction when A is applied ci sewers (2) (2) = (-f) oii ptl oth on tr Fortec en «Taster (wth eared tno tasonendat esi Teen vnc, (2) nt (a known as eigenveetors. Their directions are unchanged when the matrix A is applied to them. ‘The use of the word eigen is appropriate here, since the vectors map to a scalar multiple of themselves, WORKED EXAMPLE 20.1 Answer Let det(B a1 ‘Uset ae ttofindthe ‘character ee es Hence, -6-4+2-6 3,4 Solve the equation to get the eigenvalues, )-iG)- from which y Solving: 4= From Bx= ax, (- x+y Her diy yeeperie AE UEC ee IUCr edt lac ae -2x+y=4x For 2=4:~2x #2" from which With 22 matrices, the eigenvectors can be determined by considering a linear function. ‘This function is the gradient of the line passing through the origin that has been converted toa vector. For 3X3 matrices, this is not the same. Since we cannot represent lines correctly in 3-dimensional space, we must find an alternative approach. r24 conitertemavi A=(8 3 #).whih nance wen fm a -i2 4 ‘Then using det(A al) =0 Ieadsto | 0 2-2 2 i Use the determinant to find the characteristic equation (I~ 4)(2- a)@~A)=0. 12 4) (x) x) xt 2p tae x givesus(022)(y)=a{y).or y+ 00 3/\e z Use the characteristic equation to select values for 2 =0,y=0 and x=x. This means that x can ) yey and x Then (1-3) Now, our expression Ax x+2y-+4 When 2=1: 2y-+22=,, which shows x dy +42 Dy +22 32222 can be any value, and x will be twice the value of y. Hence, an eigenvector corresponding |. which shows that = 2y. So this time y tod which shows 42, Sox and y both, 3r=3e 4Cree Ca ae) Eten Bruise! 324 Find the eigenvalues and corresponding eigenvectors for B= ( 2 i). I 2 1, Answer From det(B — 41) the d so (3-a)|?4 Then (3 = 4)(2-4)(1 -2)- 20-1) + 4(-4 4a) This leads to 4? — 627+ 54+ One solution is 4=~I, which means 441 isa factor, So factorising gives (+ 1)(4— 3) - 4) =0, and 4. monn (| 33)6)-() For 4 From here "y, so they both depend 6 ony. Aneigenvector is (2 5 fy and then 2 From here x=, y= 2 So an eigenvector is ( 1 9,ei Smee nO Mea a Ca ur ed RUC LL uc eed En ind the eigenvalues and corresponding eigenvectors of Answer 0, Then (2- All —a?-4]-A-3)-344=0 Or 2-aa?-2-3) equation (2+ )(A- 2-3) which gives the characteristic So the eigenvalues are 4 = -1,2,3. dette From Cx=Ax weget x+y + xt2y+ It y tz For d=-k:axt yt xbly+ So -x-+2p+ Adding these gives For A= 2y, then y=-z and x For a dy +22=0. gives 2=3y and x=4y 4 An eigenvector is therefore { 1 also — Subtractin I :) ‘Set the determinant of the matrix equal to zero, Work through the algebra to find the characteristic equation. State the eigenvalues. Write down the general form before substituting in the 4 values. ‘Choose the second and third equations since they can be combined casily. State an eigenvector. ‘Start with the first equation, then use the result in the third equation, State an eigenvector. Choose the first and second equations since they can be combined easily. State an eigenvector. We now know that Ax = 2x for a square matrix A, its eigenvectors x and corresponding cigenvalues 4 It follows th ‘knowing an eigenvector and applying the matrix to it will give usfore oeed socal astonninks pon Freamitnest A= (2!) ant) ban ceensson ten ?_!)(2) (CT eres (2)=2(2) msn cismatis bas eigenvalue 2 and eos 5 4, corresponding eigenvalue and the value of k, 0-1 0\/ 1) /-2 Westart with {k -9 -6]( 2)=( &). By comparing the third elements of the -3, -3, o-1 0 1 sunmee n= ({ 1) bmsskamasamesor( 2), weit ne su 7, eigenvectors we can see that rata 1 Ineach case, state whether or not the given vectors are eigenvectors of the matrix, If they are eigenvectors, write down their corresponding eigenvalues, a aa(j 3)e=(2) ante (4) b a= (( -2)-e1= (3) ana «=(i) 6 leading to k 4, 12 + centr 4=( 1 @) ston hareftonine ear reeemtn nd Stemi ce 3 By considering M’e, and Mey, find the eigenvalues for M2. Hence determine the eigenvalues and corresponding eigenvectors of ME. © 4 Find the eigenvalues and corresponding cigenvectors for the following matrices. a4 13 * (i) » (4) -1L 3-6" 9 @ 5 themaix c=(—'s 2 4) naseigenvector ( —8). Find the value of 1-6 9 -16, p30 1 /3\/1 @ 6 Giventhat A=(_ 1 21} hascigemectors( 1), 1),{ 0), sind the values of the corresponding -1 44 Kal \- cigenvalues, as well as p and g pare © 7 sm tne shaseseiion fe hem (-£ i :) P6224 1400 8 8 3,Cambridge international AS &A Level Further Mathematics: Further Pure Mathematics 2 ay @s Se : i) a 20.2 Matrix algebra 25) i Consider the matrix A= ({ 2), which has cigenvalues 2 |. 7 and corresponding. 1 same eigenvectors as A. Letone eigenvector bey Then Bey = Ay +2, Sine and €on0t change the Girton of ey then fellows tht B oes not change destin ether, Hence, mus 48)(-3) _(-15 #(9G)-C) eigenvalue for B is two more than the eigenvalue for A Tappeat hat adding Moa matin increases tees yk Po Meni nanaeream at C= (2 33) wither. tan leer 0) ()). () 024 of Lal \2, roy (3 vow (01 8)(0)- (70) sinsanctenaest-a 02 oo) Ao, 32072) /4 01 8\( 1) =(3) evsensomanear- 02S “sp oysy. 3 tans, (11 3)(2)=(3) snsonssnaters 02 SEIN So the eigenvalues of D are -3, ~2, 3. These are all four less than the eigenvalues of C. ‘Again, adding KI to the matrix has increased the eigenvalues by &, since & was negative in Then let D= this example, If matrix A has eigenvalue @ and corresponding eigenvector e, and matrix B has eigenvalue wand corresponding eigenvector e, then (A+ B)e isequal to Ae +Be= Ze + ne. So (A+ Ble = (/ +e, as shown in Key point 20.2, 2) key POINT 20.2 If matrix A has eigenvalue 2 and corresponding eigenvector e, and matrix B hat corresponding eigenvectore, then (A+Ble= +e. igenvalue y andWee ntat 78 fT Prema a= (4 5 2) tacomatas £2545. Farah offing wo 034 corresponding eigenvalues and write down the relationship between each matrix and A. 112 1 212 1 a B=(i 5 3 bc=( 1 2 3 0 0-2, 0 0-5, 3121 173 0 04, Answer a For BA 2, 7; relationship is B= A+ 21. b For C:4=~4,~5,4; relationship is © For D:4=1,0,9; relationship is D | workeD exampce 20s | EXAMPLE 20.5 62 3 Tremasinaighonty ( Actates (i+i)e 04 > Uae Co Hani toned cine a hems A= (1 4 *) 33 a Ae+A% 20.3 Diagonalisation Contreras A= (-3 “1 nage tc nie A al ae et deal of calculation. For a 3x 3 matrix it would take even longer. Fortunately, there is a more efficient way of doing this, “3 coeponngstemccios ron be (2). ( 2) [Next we shall form two new matrices: ) hich s made up ofthe eigemsectors D~ (2 8) whic a seasing agus conn of he cna tht comand i -(3-1\f1 1\_(2 3 rotecnecasine Non secaneneuse aP=(3“!)(! 1) =(2 8) This new matrix shows the effect of each eigenvalue on its respective eigenvector. ewecasat Po, wast (18) (28) =(-2 8) tocencheenecris multiplied by its own eigenvalue. So, for t 102 sovcomiesn=(1 3 8) tismatichsegenalis 1=1,3.4 acre 00 4, 0, \ 2 ()G SG po? tas ma(-11 2) whos ein eigenen eat 03 0 10" 8 80) wherein. gon os esa corresponding to their respective eigenvectors.coy nt Cee ERnO Se Tee eum en ee Luu ed Potty Let A= (: %): where the eigenvalues are 2,2, and the eigenvectors are (%): ¢: ‘) Investigate, with this general case, whether ornot AP = PD. This can be extended as shown in Key point 205. (ete If mateix A has eigenvalues 2.43.25, dy and corresponding eigenvectors \ fos Apne eee) n) fs eee eee) sf fo |, then the matrix D= {0 0 2... 0 | and matrix wl \yet \sy 00 0 wd fre such that AP = PD. Hence, A= PDP~!. Note that P HOO OY Py Oe ee Ome eke Coeur 004 0 0G ayn. 0 | so itis easy to find powers of D. HHO say ORG Oe e Ae) Any matrix that can be written in the form A = PDP" | is said to be diagonalisable. Tomake us ofthis eltionship, consider the matric A= (}, 2), Cones for example, that we wish to determine A®, [f we use the form A=PDP-! then A™=(PDP-'), which can be written as PDP~! x PDP~! x ... x PDP~!, Note that all internal products of the form PP are equal to I. So A®® = PDDD ... DDP~! > A” = PDP! cae vector (2). 1 andcoespnding mses (!).(!) You can review basic matrix operations in Chapter 4 of this book. The matrix A has eigenvalues 4 sop= (8) snie=(' {) wesortsinarearmmanansor = (2 ]) nc a*=(1 CB I)°C )=C)06 2) 2) aerremnnnCar pn cod icant 120 A matrix is given as B= (: 2 ') Itisknown to have eigenvalues 4= 1,2,3 and corresponding eigenvectors 002, 1) /2\ (0 o}.( 1).(1). rina we of \-1) \o, Answer 100 1 2g Le D=(0 20] andP=(0 11). 003, 0-1 0, ‘Then performing the row operations ry-rs +13, Nr) = Bayh) + Pre ary — Ps ON the augmented matrix 10 0} Sowith Ptand D'=(0 64 0 1 728 ‘) aes ICES a Fry 1 Bis (op fag 0 11 ; 010) leadstothematrix P'=(0 0 - O-10:001 ol 1 0 07 1o21\fl 0 o\fi-1 1 weeannowsaythatB°=(0 1 1 }(0 64 o}(0 0 = 0-1 0/\0 0 m90/\o 1 1 wish voisousoe (09 es 0 0 64, 2-30\ (5 Given that one of the eigenvectors of the matrix A=(1 a 3) is( 1 ), sind matrices P and E such that o os) \o AS = PEP™!, (You are not required (o find P*!) ro 2-50\/s\ (5 startwith (1a 3)(1)=(sie 06 Sho) Vo Hence, 4;=1 and a jaa -s tim | 1 4 fo 6 equation (2 a4 ~ 4)(5 — A) + 5(5 4) =0, Then (5 ~ayf(2~a)(—4— 2) + 5] =0. Simplifying gives the equation (5 ~ 2)(2 + 3)(4- 1) =0, Hence, 42 =~3,43=5 a5 ax * 1 hat the othe! Then Ax =Ax > x —4y 43: Saee CnC ROS ee aur ca un LC ed 2x5 Look for the one equation that explicitly determines one of your values. v 0,y'= x und so an eigenvector is () ) Combine results ifnecessary to obtain the 0, eigenvectors. —5y=5x When 1=5,x—4y +32 5y, giving y= b 1s x, Hence, an eigenvector is (_ —9 sou Po( a) hae E=( 01 0 Note that AS requires D’, which is denoted 00 312 by E. State P formed by the three eigenvectors. ) enn i It was stated earlier that a matrix written in the form A = PDP* is diagonalisable, From this expression, the only matrix that might cause a problem is P~!. If P~! does not exist, then our relationship does not exist: A can be diagonalised only if P~! exists. 231 Consider the matrix A= ( 2 ‘) Ithas characteristic equation (1 ~ 2° — 1) = 0. 901 ‘Any matrix that is rnon-square does not have an inverse. This implies that non- square matrices cannot be diagonalised Since there are only two eigenvalues, there are only two different independent eigenvectors 1 a row These are ( 0} and ( -4 }. Sonow P 0 1 inverse, So the matrix A is not diagonalisable which cannot possibly have an ectacueeies Show thatthe matrix A= (J =) isnot dagonalisable Answer Using det(A ~ 41) Use the determinant to find the characteristic equation characteristic equation is #2 +224 =0,or (A+ 1)=0. So A=-1 Note that only one value of 2 exists, Hence, there is only one distinct eigenvector, So P is ‘Only one eigenvector means P is not an not a square matrix, which means it has no inverse. nxn matrix. ‘Therefore, A is not diagonelisable. Hence, we cannot form A= PDPaoe acces Beast e 1. State whether or not the following matrices are diagonalisable. 3-2 11 =10 -9 °GH) 8G) CR) @G® 2 Giventhat the matrix (5 4) isnot diagonalisable ver real, find the values of @ 3. Given that the matrix @ ) is diagonalisable, find the smallest positive value of k, where k is an integer, that gives integer eigenvalues. 4. Find the values of the following matrices. a (34) » (37) 02, -4 6, @@® 5 Find which of te folowing matics are diagonalisabe 102 1-0 30-1 2 (083 » (34) e (ota 201 9 02 20 0, 339 vf 1\ fi @ 6 Themarix a=(1 5 0) naseigemestors(-1),(-1),(1). rind ee 004 of Va) to matrices P and H such that B'= PHP™!, where B= A — 31. ki? QO 7 Focthemateix a(: 3 8) nt eines tiger stor foe te ook anes hin =O sod E= 2 apis. why te matrix & cannotbe lagonatiecd when eo > 1) te theses dds 2 Use the leading sliagonal 20.4 Systems of equations Qv-y+32=4 Consider the system of equations 3x 4 2y + 8: = 13. We want to find a solution for x,y,z 4x4 2y4 lz 16 2-1 3\/x’ 4 First, rewrite this systemas (3 2 8 }(»)=(13 4 2u/\2/ \v ‘There is a very good way of approaching this using row reduction methods.ei a aE INOS eae uC eds Ce uC Lod 2a3: 4 reeteatmeneima (32 #8) teopmtim nn 2nrman— tia ite 213: 4 and ry-+7r—4ry willgive (0 7.7 : 14). From here we have 0 07: 9, and, hence, = So the system has a unique solution (3,2, 0). Procionuatsrer) Dv +52 Find the unique solution for the system of equations + y-+ xoy+ Answer “(10 mt Create the augmented matrix reer aupnenet mais apply the row operations 1~2r; ay ga. a getthematrix(G 2-1: 3 Apply operations to get row echelon form. 002; 2 Hen -Lyelx 1 Solve for a unique solution. x+2y+32=1 What happens if there is no unique solution? Consider the system 3x +4y + 132 4x4 Ty 41 12 3\/x\ (I First, write this as (3 4 13)(y)=( 5), so the augmented matrix is 4714/\eJ \s 12 aif 3.4.13: 5). Perform operations rry—3ryrycrs—4r1 and 154275 ~ ro, giving 4714: 5, Weo@eae cecal o24:2 0 00: 0 123\f/z\ fl In Ax=b form, thisis (0 -2 4}(y)=(2 }, which gives 0 oo/\z) \o, -2y +4; Simplify these equations to give y= —1 +22 and x nor istesnermicr Weonweensenstines (1) =(3)+(2) DAG Note that, since both and y depend on 2, we can introduce this parameter into our system, ‘As zisa free variable, it is free to change in value, This means that there is an infinite number of solutions,eerie etc istuu tree) Det 3yte Find a solution set for the system of equations 4+ 10) + 5: 2x My +72 Answer @ 4] The augmented matrix is (4 10 5 ‘Write all the coefiicients in an augmented 27 matrix. The row operations r-+ry Apply row operations until the last row Bat ey cannot be altered any further. aive(0 43 000: 0, ‘Write down the equations that relate two variables against the free variable, State the solution. Ifa system of equations is given as 3x + y+ 37 =2, can we determine whether or not the 6x — dy +91 system actually has any solutions? 3 -14\/x\ (1 Starting with (3. 1 3}{»)=( 2) weave the augmented matrix 6 4 o/\:) \3, 3-450 613-2 6-49; 3 With the row operations 5-73 ~ 2ra,ry+r3 — ry 75-475 + 37a the augmented matrix pares wom: (32 1/1) noses esters 02. ofsms hi 030} (recanted 1 2 has no solutions. 3 ok dy = 22 ‘Show that the system of equations Answer i 45 Startwith (1 5 0 313-4 ‘Use the augmented matrix with row operations to get to row echelon form.Cambridge international AS &A Level Further Mathematics: Further Pure Mathematics 2 Inconsistent values mean no solutions. Ifwe have the system 2x-+5y+2=9 and we are to interpret these systems of equations, x+3y432=10 1 fia performing row operations will reduce the augmented matrixto(O 1 3 7 7 oo 1:2 Since the bottom row has a distinct solution for =, we can see that there is just one unique answer, The three equations can be modelled as planes, and the unique solution js the one point where the planes meet. xtayt2=1 For the system 2x+4y-+32=3 we perform row operations until our augmented xt Ty +d 1 aaa f matrixis ofthe form (0 -2 -1 ; 1). This system has an infinite number of 0 0 0; 6 solutions, so all three planes meet on a line of intersection xt dye Finally, consider the system _x-+3y—22= ‘When we perform row operations on the nldite ps ed svennsimacwceaneroat (0-1 {1 weacno asin : oo 02 ‘The diagram on the right shows one example of three planes without a solution. =z ; © 2 Using algebra, determine whether each of the following has unique solutions and, ifso, state those solutions, a 3x-y b m+ © Sy- dyad 6x +2y=-13, Br 10x -8y=4 2 @ 3 x\ 5 @® > vie =(2_$) and mann (3) ns someing axe, ytne x= (2) tu ‘a and f such that there are no solutions band such that there are an infinite number of solutions ¢ aand P such that there is a unique solution,arrnce @ _ 3 Determine if the following systems of equations have a unigue solution, an infinite number of solutions or no Solution. IF there isa unique solution or an infinite numberof solutions, calculate the solutions xt4y~22 Qetytz=! a -2x- 10y— b ax+3y 4102 c 3x4 14y 442 ax x-ysl © _ 4 Forthesystem ofequations 2v—y-+ Sz =4, find the value of & such that there is an infinite number of solutions. x44 1S =k © 5 Three planes are given: xa 4y442=3 x-Ty+llz=4 De Syt3e05 Find the unique point where all planes intersect. x= 6y=2 © 6 Forthethreeplanes 2x-+4y — 172 =4, find the point of intersection, 3x4 12-337 =6 xty+3e 7 For the system of equations x ~ 2y + 2z = -1, state the number of solutions when: 3x46) +az=6 aa 3 » © a=10,b= @ RIESE ‘Show that if is an eigenvalue of the square matrix A with eas a corresponding eigenvector, and jis an eigenvalue of the square matrix B for which e is also a corresponding eigenvector, then 2+ 4 is an eigenvalue of the matrix A+B with eas a corresponding eigenvector, ( 3-1 :) A=(-4 -6 -6 5 11 10, PDP* Find a matrix P and a diagonal matrix D such that (A +B} Cambridge International AS & A Level Further Mathematics 9231 Paper 1 Q3 June 2008 Answer Start with (A + B)e= Ae+ Be, which is e + ye = (2 + ye 1 3-1 0\/1 1 A(-1 )=(-4 -6 -6 }{ -1} =4( -1) gives an eigenvatue of 4 1 su 10/\ 1 1Cambridge International AS & A Level Further Mathematics: Further Pure Mathematics 2 Checklist of learning and understanding For eigenvalues and eigenvectors: © To determine a characteristic equation, use det(A ~ Al) © To find the eigenvectors use the relation Ax = 2x. ‘© Eigenvectors do not change direction when a matrix is applied to them. For matrix algebra: Given that Ae= Ze and Be = jie: ‘The matrix A+B has eigenvalue A+ and corresponding eigenvector ‘The matrin AB has eigenvalue Zu and corresponding eigenvector e, ‘The matrin A+ KI has eigenvalue 2+ and corresponding eigenvector e ‘The matrix A’-+ A+... haseigeavalue 342%... and corresponding eigenvector e For diagonalisation: ‘eA matrin, A, that can be written in the form A= PDP~ is said to be diagonalisable, Hiss 0401, ose oa 0 0 ‘The diagonal matrix D= {0 0 2 ... 0], where each eigenvalue is placed in the oo00 i An leading diagonal ofa square matrix “The matrix P has its columns made up of eigenvectors that correspond tothe eigenvalues in D. Af there are fewer eigenvalues than the dimension of the matsx, then there will be insufficient istinet eigenvectors, This means that, in general, the matrix P cannot have an inverse, and so the matrix A cannot be diagonalised, There are exceptions to this but they are not covered in this course For systems of linear equations: ‘© Fora system of equations, where the reduced augmented matrix is of the form (: mel *) ets omer doa: Dy + If a@=0 and f=, then there isan infinite number of solutions. © If @=0 and £40, then there are no solutions. © If a0, then for all BER there will be a unique solution, For inverse matrices: (© Fors matrix, A, with characteristic equation P.4(2) = ai! + Ai? + ch-+ d=, the equation aA} + bA24 cA + dl =0 is also trueree Caney eT 3-1 0 @ 1. Find me cigenvatues and corresponding eigenvectors of the matrix A= (1 2. — 0-1 3 Find a non-singular matrix M and a diagonal matrix D such that (A ~ 21) =MDM7!, where Lis the 3x 3 identity matrix, Cambridge International AS & A Level Further Mathematics 9231 Paper I Q9 November 2010 10 2 nthematrix A=(0 2 0 O11 Find the characteristic equation of A. 2 Youare Hence, of otherwise, determine the unknown constants for A? + aA? + BA +71 Hence, or otherwise, find AW! 0\ /-1 1 i the matrix A, fi A™, where m is a positive integer, Cambridge International AS & A Level Further Mathematics 9231 Paper 13 Qi.1a June 2011pyiacyacyalarel eda) In this chapter you will learn how to: @y ons for — in cas the relation between and y is defined both implicitly dt ‘bolic functions and inverse trigonometric for a functionCee Pree Cee tes Where it comes from ‘What you should be able to do AS & A Level Mathematics Differentiate functions such as Differentiate the following Pure Mathematics 2 & 3, Chapter 4 | e, acos(f(x)) and a la f(s). functions, a xe | b 2sin(x? +1) | « AS & A Level Mathematics Differentiate functions implicitly a Pure Mathematics 2 & 3, Chapter 4 Ba et eee xy-y wee, AS & A Level Mathematics Differentiate functions defined dy. Pure Mathematics 2 & 3, Chapter 4 | parametrically, Bid) for thestiot parametric equations What else can we do with differentiation? In this chapter we shall look at the differentiation of some new functions, such as hyperbolic functions and inverse functions. We shall extend the implicit and parametric differentiation techniques you learned in your A Level Mathematics Pure Mathematics 2&3 course ‘We shall also study Maclaurin’s series for a variety of functions. This will allow us to approximate functions with a polynomial. Differentiation is the section of calculus that looks at rates of change of variables. It is used in many different areas, such as Physics, Chemistry, Engineering, Medicine and Economies. Calculus was developed by both Isaac Newton and Gottfried Leibniz. It has become a pivotal part of modern science and engineering. 21.1 Implicit functions ‘You will have met implicit functions within A Level Mathematics Pure Mathematies 2 & 3, Chapter 4, where you will have found only the first derivative. In this section we shall be extending this to incorporate the second derivative. Implicit drferentiation allows us to differentiate functions that are not explicitly writen as = (x). For example, consider the function 9? = x, Differentiating both sides with respect dy oa dx 2y 4 tox, wecan write 2p a ‘When differentiating y, think of this as 4. (), as shown in Key point 2.1 Ths the same Ie 4 ay as © ityour function is, for example, 2x -+ x (ny) and then diferentiate. This is also an application ofthe Iny, do not start with Instead, start with 2o7x+y) a chain rule.a cen O Se ee a aur eat uaa uc oe (RT eres To diferentiate, start with )= instead of = and then differentiate, Consider x} +»? = 2y. Suppose we want to find the first and second derivatives with ‘ ay _ a “(2y) and so 3x? + 2y a Pac " dx dx respect to x, We write Sesy (1), Rearranging gy 2 ives us ©” = 2% .(2). To determine the second derivative, itis general givesus G95") idea to consider (2), since differentiating this result would mean having to deal with a quotient. Instead, we differentiate (I), the original form of the first derivative: é ay ay ‘(3% x *) £02) & dx) 7 ae a ae ey _&(5) ace ay Using the previous result, which simplifies to ay _6x(2- 2y) + 18x" ae @-2 Picante Given that e* Iny, find the first and second derivatives with respect to x, Answer Find the first derivative. Differentiate the first derivative in its, current form to get the second derivative dy__yet & So and a Rearrange both results, dx 1 —2ye! ae y-2re i oy. ae Using the result for $Y, rewrite the second derivative as Substitute SF into the result for the second ree saren( erivative @ 1=2y x1 = 2ye*) Fy _ yell rye" eX + ay7e™) te? oe i Simplify as required. ae Obtain the result, or any equivalent form.Ca ese tr We shall now look at a slightly more complicated example: x3)? + y = sinx. a ay Differentiating, 3x2y2-+ 2242 4 2 Wet ag ae cos.x. Notice that we now have a triple product, a 2x8y fe In order to deal with this we need to see how to differentiate a triple product. If y=uvw, where «= M(x), ¥= g(3), w= h(x), then y + By = (u-+ 6u)(v-+ By)(w + 81), Multiplying out the brackets leads to: 2+ By Sune + uBoBw + y6uBw + wEUBY + wBw + snBy-+ ywBU + Bub vBW Notice that y= sw will now cancel, We then divide through by x to get: oy Ww, rOudWw wEuby | uvbw | uwSy www SubvSie Su, wa, Waban hon i end to ox bx Bx ‘This means that the terms “©°5", WHY ig ub are all small enough to be ignored Ce So P= ww H+ 1 2 + 5, as shown in Key point 21.2 a dx dw dx dy’ : ane 2) Kev Point 21.2 {fa function is of the form y= ure, where each factor is a function of x, then the triple product. &y pl ifferentates to give =» + se TE gy Mt oe aie ax ax dx 3,4 dy In our example, we got to the point where 322y? + 2x4 = 7 = cosx. dy dy ‘dy\? Difereniating again gives 6x? + 6x3y P+ 6x3y 2 4 2092) 4 avy hy SE sins pingagain gi ae rr Pax dy, 4 "a @ & Wecout arong the sete find resins for and 2 resctr ey An implicit equation is given as xe**” = (x + 1)?, Find the first and second derivatives with respect to x: Answer oy Begin with e+? + x(1 if ee =%x41). Differentiate once. ay a &y 5 ea Thea (: + ew + ( 32 er + xere2 Note the triple product when differentiating a i he & a second time, x(142)rerr (43) dy _ 2x41) exer? 80 5 Sn ee nite ishdetratie State the frst derivative, aan : And S225 The algebra has become too complicated to write sesenil incl fis the second derivative just in terms of wandi MeO Te Gnu ra Cu ena CLE uC Discuss in groups how you could attempt to differentiate the functions y= x" and on Deal with frst Looking at Worked example 21.2, there is a good reason why we did not simplify the second derivative. If we only need to find numerical values for the first and second, derivatives, we do not need to simplify the expressions. Consider the funetion x2 +.xy° = (y+ 1)', where itis known that when x= 1, y= —1. To find a ey numerical values for © and ox we start with the first derivative. 4 21 <0, whieh & aye 2a So 2x4 2aySh y= 30+ 1)PG Then with x= 1 y= ay 4 ly gives us 7=3, Wedo not need to find explicitly as we only need its mumerical value Sorw+6(2)o+n oy (a2 Ditereiningagan, 24424 20(2)' 42 & S ax, roccianuatere Given that y? + yx? =e" passes through the point A(0, 1), find the values of the first and second derivatives at the point 4, Answer Differentiate once. oad First, 39°24 y ” ar > + 2yx a ay Using x=0 and y=1 weget 3 Use the given values. There is no need to rearrange to 3 find > explicitly. Differentiate again. a Use the given values and the resutfor Determine the value of the second derivative.Breen e6e060 ry forse ery ay For each case, find the first derivative“, first & b yrae © tan@+ y= ay Find the value of the second derivative © of ae ax ye" + y? at the point (0,0) Find the value of the first and second derivatives of In (x + y) = 2y at the point (1, 0). Given that xp = sin( + y), find the first and second derivatives with respect to x. C, is defined as y= In x+.e” ~ 1. Given that C passes through the point (1,0), find the a values of and < at this point ae ae ‘An implicit curve A curve is given as xy" = f, where a, f are constants, 4 & ® a Fing © b show thar 2% 20 +9), ae ae ae The curve Cis given as (x + y)*= x. Given that the curve also passes through the point (1,0), determine the ya values of and at this point, dx dx? v3 An implicit curve is defined as sin2xcos2y =~, It is known that the curve passes through the point ° *) Find, atthe point P 36, dy ay pad b = dx act ‘The French mathematician René Deseartes worked fon the now famous curve x? + y? = 3axy. In 1638 Descartes challenged Fermat to determine the ‘tangent line to the curve, Fermat did this without any knowledge of calculus! Once ealeulus was discovered, differentiating this function was a much easier way to find the tangent.eee nr IOS Te Ua CMC ta Ferd 21.2 Parametric equations You have met parametric equations in your A Level Mathematics course: two funetions are depen (0) and y= g(i), where tis the parameter int on a common parameter, for example, x ‘To determine the gradient of a curve represented by a set of parametric equations, we must ox ana first determine SF and FThen, using the chain rule é ‘Another way of writing this is 7 = (5) x $, where ys iferentited wit respect tot and then multiplied by *, dx ay ax 4 Lythen 21 and a at au Notice that the first derivative is a function of t, so to determine the second derivative we For example, if x=? and ay rust differentiate 5 with respect t d ay) de (2) xh asshown in Key point 2.3, Notice that the fist and second ax) * ae a derivatives are obtained using a very similar approach, Crone For the parametric equations x= {(0), "= g() the fist and second derivatives are given by dyad de eee aya e) at ] a alee ox Going beck to our example with x= 2,y= 1+ 1, we have already found that ay 1. For the second derivative, first find $(4) 3a (4) 2 ww \2r = ‘Then using ae 2 a 2ar\t : @ dy) d wy dt\dx) “dx dt 2ees ery Ene peter) ?, find the first and second derivatives of y with respect to x. and y Given that r+ 201 + an(1 + 0). This simplifies to -2(r + 1)(3r + 1). dy ‘dy Then £( xe A) x8 1)x (4 1, aide) * ae 7 2+ DG + 1) x0 41 31+ I(e+ UP. Perens dy @y Ifx= 1 determine the functions ~ and —*, dx dx? Answer First, 237-1 and % e's re'=e( +2) Consider the curve represented by x= cost,» find the coordinates of all the stationary points, and to determine their nature, dx. ay oy é 2eos2r ~sini, = 2c0s2r and, hence, aa mee ae sine The first step is to obtain 4 ‘Then for turning points we need z = 0, or for parametric-based derivatives, we just ay need a So let 2cos2t= 0. This leads toeit curd nO eran Ua Cau en ae aCe ‘To determine the nature of the stationary points, fitst find aps int(~4 sin21) ~ 2cos (cos #) as\ax) ~ , then multiplying by 1 Teads to dx sin? 2(cos2tcoss + 2sin2¢sin?) z sin (4) wh =850.Hence, we have minimum at ( wa Eicon Btu ( ~ 3, Find the coordinates of any ‘A parametric curve is represented by the pair of equations +4 stationary points, and determine their nature, Answer . 44> dy 32-60 Begin with 22 = 47+ 1,22 = 37-61. Henee, 2 = 3°=6 Find the derivatives ofeach a a dx ae +t variable with respect to ¢and determine © mine ©. & ay & For turning points we need "= 0, s0 3/°— 61=0, giving 1= 0,2. For turning points let d = 6(12F) _ 483 — 124+ 6r-6 Determine dy) _ (P+ 61-6) - G Y ( )S WGew= Differentiate with respect at\ax Ge+1F a a) at aa tes) a & 2 When 1=0,°% =~6 <0. Hence we have a maximum point when 1 = 0 Since < Owe havea ee ‘maximum. When 1=2,@%=-2 > 0, Hence we havea minimum point when since “> 0 wes en 12s ave @ minimum pe ince T5>0 wehavea minimum, Determine the coordinates of both stationary points. Also, when 1=0,x=0 andy =0,and when ¢=2,x=18,y So at (0,0) we have a maximum point, and at (18, ~4) we have a ‘Write down the coordinates of minimum point. ‘each stationary point and state whether it is a maximum or minimum.Ce Teen) 7 © + Foreseh case strane est orate etn obtain heseconddeatne © @; © 2 Find the value of the second derivative of the parametric equations x=, 122 a2 4.athepeint hee yy © 3 Find the vatues of & and a ofthe parametric equations =e" y at the point where ¢= 1 2 =32 find the vatueot © when ¢=2, ae © 4 Giventnat x @G® 5 Aparametric curves defined as x= In(e+1) and y a Find y in terms of 7. ax b Show that the second derivative with respect to x is always positive. P+ 1,y= Ine is valid for ¢> 0. @ 5 tpnsnans a Sho tht hereare nm respi ea ee lo te LISI gi PQr+ 1p © 1 it vasintey cos isvali for 0:¢ 1 < find the second derivative wth epost toxin ems of @ e Given that x 244 find: 7 2 the value(s) oft for any stationary points) b the nature of any turning point) is valid for allt @® 2 The parametric curve represented by x= e+ e-%,y 4 a Find © and state the number of turning points et 0 b Show that the second derivative with respect to xis — 21.3 Hyperbolic and inverse functions ‘The hyperbolic functions were introduced in Chapter 19: © cosh, © tanh,ent a cnet ROSE Tea UCC or aa LUC -oshx Starting with » = sinh =© Differentiating y= cosh = sinh.x. These relationships are shown in Key point 21.4, ‘£) Key POINT 21.4 ‘ sins, then é oshx. i a cosh, then = sinh. & ¢ y=sinh’x da Use a substitution to simplify the funetion. u_y ax ay 2cosh2x, Use the chain rule to get & ssh u. Again, make use of a substitution. du _5. dx ay So = 2esiahy From the chain rule we get the result ¢ y=(sinhx?? 4 So P= 2sinh xeoshs, oF sinh? ‘Treat this as (2) finishing _ iy with Fe nf ()tcort and use sinh 2A =2 sinh A cosh A. 4 d 2 _ ssinh 3ysinhdx + 4cosh Ixcoshdx. Differentiate using the product rule and chain rule. Notice that hyperbolic functions behave in a very similar way to trigonometric functions, Let us consider the derivatives of tanh x, sech x, cosech x and coth x.Crees eect? sinhx ae inh xsinkx ay iy then, using the quotient rule, © cosh ne is aa 2 ty : Using cosh? — sinh? = 1, 9 = mechs If y=sechy= YY coshxXO— Licsinh ot ctan ‘cosh’ dy cosh? 1_ 4 | sinhxx0— 1 x¢coshs I y= coseche= "SAX = ~cosech xcothy. sinha ax sinh?x coshx dy _ sinh.xsinhx~ cosh xeosh x If y=cothx = 284%, 28 Ayain, using cosh? ~ sink? Fee Siahix’ dx sinh? x = : ay = ~cosech?.. ds ‘These results are shown in Key point 21.5. (Tare a Iy=tanhx, then & & y= sechs, then 2 = -sech tanh a If = coseeh, then 4 If y=cothy, then Trcsn ae a) 4 Find [> foreach ofthe following functions a y=2xvsech(3x— 1) b y= In(tanhx) © pa xtenhe ds sinh(x + y) Answer 4 a a = 2x sech(3x ~ 1) + 2x x (-)3sech(3x — I)tanh(3x~ 1) Differentiate and take out the common BS factors, This simplifies to 2sech(3x ~ 1){1 — 3xtanh(3x ~ 1). b Using the chain rule we have Recall the method for differentiating : ai logarithms from your A level Mathematics om -- course. dx tanhx ~ coshxsinhx’ °" dx sinh2x ‘Also make use of the double angle formula for sinh24. sier anne nO ee Gta Ee uM Coed Using the chain rule Using the product rule and chain rule. & ere sear 2x x eB +27 x (—cosech? x}e rr 08 Je This can be simplified to xe*(2 — xcosech? x). Take out common factors. aw seis 4 aw Zeer Differentiate implicitly. cosh(x +) Feed Rearrange and state de cosh(x +9) = We shall move on to the inverse functions of trigonometric and hyperbolic functions, Consider, for example, the function y x. This is also known as y= aresin x. We want to be able to determine the derivative of this function, as shown in Key point 21.6, How do we begin? Since we have already met implicit differentiation, and we know how to differentiate the sine function, let us first take the sine of both sides. 1 - <1. since cosy=vi—sin?y wecan ax cosy SOF NL +V1— =. Then from the diagram we see that the curve of y= sin”! x ay ive root and — z oy So siny = x, then cosy >= 1, and s y > 0 say that cos; is always increasing. Hence we take the pos!ET ee Similarly, for y p Sowith 2 = we look at the denominator. Since sin: t follows that cr siny ay siny = 2¥VT= From thepraph we can confi that © = 1, eg shown in Key T= point 21.46 Wecntate ey Take tan of both sides. Differentiate implicitly and use ia trigonometric identities. Given that 1 + tan? y = sec? y: a i dx 1 +tan?y ay & so 2. By dx 14x Stee de If y=sin's, then a Vee ay 1 If ymcor's, then = 7 dx 1 If y= tan @ a, then 2 a iseCambridge International AS & A Level Further Mathematics: Further Pure Mathematics 2 Prciecuatssec) Find the derivatives of the following functions. a y= xcos Answer oy & standard result for eos’ sides State the result c implicitly. Differentiate asa product and quote the 3x,and then Sinee we have 3x itis best to take tanof both ‘After taking sin of both sides, differentiate We have seen how to differentiate inverse trigonometric functions. Now we will differentiate inverse hyperbolic functions. Starting with He sinhy =x. Then coshy©2=1 30 Fm Us Pose ee nian Yas | 8° ay coshy the result cosh?) sink?y= 1, we have coshy = VT# 2 From the following graph itis clear that the gradient of sinh“! is always positive. So ey © 1 __. as shown in Key point 217. oe Vie =sihs ) za We itty — ay Next consider y= cosh!x. Writing it as coshy = gives us sinh y Again we use the identity cosh*y — sinh?y = 1, giving sinhy = Va" =Chapter 21: Differentiation From the graph, we see that the function cosh~!x has a positive gradient for all values of x éy ints domain, So 2 = Ee Penatrtees Find the gradient of the curve y = tanh"!s. Answer Start with tanhy Use implicit differentiation. ‘Then using sech*y =| ~tanh?y, ‘Then use the identity to rearrange the result. oy we can say that L cam say that = State the derivative, x 1-2 2) key Point 21.7 dy ioh"'y, then a If y=coshts, then 2 yecosbrts, then ty y rans, then ae Banat © 1 Bydifferentiating y- 1 dy 4 Te, show that 2 ease, at 2sinh 2x. © 2 Giventhat » gy. nb (2x 43), find 2 HHQx +3), Find © 3 Find the first derivative of y= sin"! (x).ie eR OSe Te aur eee he CLE aco © 4 Differentiate the following functions with respect to x. Simplify your answer where possible. b p=xsinh(2x? - 1) d potan'(r+y) dy © 5 Givenshat y= sins, fn * w @ 6 Acurveisdefined as y= cosh’x’ a. Find the first derivative with respect to x, stating the number of turning points. b_ Find the second derivative with respect to x @_ 7 Establish a result for the first derivative with respect to x for the following functions, a ay=sin“! bx b ay=tan bx c d ay=coshetbx e ay=tanh'bx © ® Acum a. Find the values of ¢ where the gradient is zero. , C, is defined as being x= 3tanbr"s, y ine, where -1.<1< 1 b Find the second derivative with respect tox when 1= 0. 21.4 Maclaurin series B You have already met binomial expansions in your A Level Mathematics course. The expansions that are particularly important involve approximating a function using a polynomial ‘Consider, for example, the function a Lextat+so+.... Provided [x|<1 this approximation is valid. To get a better approximation we want x {o be as close to zero as possible. But what if the function we want to approximate is not of the form (a+ bx)"? For example, we might want to approximate y =e ‘To do this we are going to use a special case of Taylor series. The Taylor series of a function f(x) is given as fa) + (@e-a)+ where f(x) is differentiable an infinite number of times, and the con =a at which we evaluate the function. point ‘We shall work with a special case of Taylor series, where a= 0. This is known asa PO fOr + ()+POx+—— + Maclaurin series, where £(2 = 00) The Maclain series can also be epresented in summation form as > shown in Key point 218, where (0) is the mth derivative evaluated at the point x = 0Chapter 21: Differentiation (Sines ‘The Maclaurin series for an infinitely differentiable function f(x) about the point x = FO? fOr meat is given by £6) = 110) + POx+ How does this work? Consider the funetion f(x) the expansion, Then f(x) =(1— x)" gives (0) FG) = A- =3)?= (1-3) gives £0) = f(x) = (-2)(-1)(1 — x)? = 2(1 = x)? gives £"(0) = 2. £"(x) = 2(—-3)(-1)(1 — x) 4 = 6(1 - x) gives f"(0) = 6. P(x) = 6(—4)(—1)(1 = x5 = 24(1 = 29 gives £9(0) = 24, 2a, 6x! | dat a ar at Note that this is the same as the binomially expanded result {for which we have already found So flx) 12+ which is 1+ x4 32+. 4.x! as before, Let us find the Maclaurin series for f(x) = e* about the point = 0. £(x) =e%, giving (0 a P(x) =e%, giving £(0) =1 e% giving £"(0) & giving £*(0) PO) 1%) =e%, giving 0) =1 1x) =e% giving (010) =1 oo # Soetatexee ‘The following diagrams show the Maclaurin series for e* fitting more closely to the curve fix) = e* as we add more terms. 3 £%O)x" In the summation form —*., we are considering what happens as n increases.