® INDIA EDITION DIFFERENTIAL EQUATIONS any RESTRICTED! FOR SALE ONLY IN Shepley L. Ross—__ DIFFERENTIAL —— EQUATIONS ~~ ‘Third Edition i Shepley L. Ross Unicity of New Hampshire WILEY- INDIA| | | | | DIFFERENTIAL EQUATIONS Third Eattion Copyright® 2004 by John Wiley & Sons, Inc. All rights reserved Authorized reprint by Wiley india (P.) Lid, 4435V7, Ansari Road, Daryagan), New Delhi 110 002, All rights reserved, AUTHORIZED REPRINT OF THE EDITION PUBLISHED BY JOHN WILEY & SONS INC., UK. No part ofthis book may be reproduced in any ‘orm without the writen permission ofthe publisher. Limits of Liabitty/Disclaimer of Warranty: The publisher and the author make no representation or warranties with respect to the accuracy or completeness of the ‘contenis of this work and specifically disclaim all warranties, including without livtation warranties of fitness for a particular purpose, No warranty may be ctoated or extended by sales or promotional materials. The advice and strategies contained herein may not be suitable for avery situation. This work is sold win the ‘understening that the publisher is not engaged in rendering legal, ‘accounting, or other professional services. if professional assistance is required, the services of a competent professional person should be sought, Neither the publisher nor the author shall be {iabie for damages arising hereirom, The fact ‘that an organization or Website If referred to in this work a8 a citstion andlor a potential source of further Information does nof man that the author or the ‘publisher endorses the information the organization or Website ray provide or recommendations it may make. Further, readers should be aware that internet ‘Websites listed in this work may have changed or disappeared between when ‘this work was written and when itis read, Wiley also puitishes its books in a variety of electronic formats. Some content that eppears in print may not be avaleble in electronic books. For more Information about Wiley products, visit our website at wnaw.iley.com. Reprint : 2010 Printed at : Rajiv 800k Binding House, Delhi ISBN: 978-81.265-1537-0== PREFACE———_— ‘This third edition, ike the first two, isan introduetion to the basic methods, theory, and applications of differential equations. A knowledge of elementary calculus is presupposed, ‘Thedetailed style of presentation that characterized the previous editions of the text hhas been retained. Many sections have been taken verbatim from the second edition, while others have been rewritten of cearranged with the sole intention of making them clearer and smoother. As in the earlier editions, the text contains many thoroughly worked out examples. Also, a number of new exercises have been added, and assorted ‘exercise sets rearranged to make them more useful in teaching and learning ‘The book is divided into two main parts. The first part (Chapters 1 through 9) deals: with the material usually found in a one-semester introductory course in ordinary differential equations. This par is also available separately as Introduction o Ordinary Differential Equations, Third Edition ohn Wiley & Sons, New York, 1980). The second part of the present text (Chapters 10 through 14) introduces the reader to certain specialized and more advanced methods and provides an introduction 10 ‘fundamental theory. The table of contents indicates just what topics are treated. The following additions and modifications are specifically noted. 1. Material emphasizing the second-order linear equation has been inserted at appropriate places ia Section 4.1. 2, Newillustrative examples, inciuding an especially detailed introductory one, have been written to clarify the Method of Undetermined Coefficients in Section 43, and a useful table has also been supplied, 3. Matrix multiplication and inversion have been added to the introductory material on linear algebra in Section 7.5. 4. Additional applications now appear in the text in Sections 33 and 72. 5. Section 7.6isa completely new section on the application of matrix algebra fo the solution of linear systems with constant coefficients in the special case of wo ‘equations in two unknown functions. The theory that occupied this section in theiv vestace. second edition now appears in Chapter 11 (see note 9 following). We believe that this change represents a major improvement for both intcoductory and inter- mediate courses. 6. Section 7,7extends the matrix method of Section 7.6to the case of linear systems. with constant coefficients involving m equations in n unknown functions. Several etailed examples illustrate the method for the ease m = 3, 7. Both revised and new material on the Laplace Transform of step functions, translated functions, and periodic functions now appears in Section 9.) 8 The basic existence theory for systems and higher-order equations, formerly located at the beginning of Chapter L1, has now been placed at the end of ‘Chapter 10. This minor change has resulied in better overall organization. 9 Chapter 11, the Theory of Linear Differential Equations, has been changed considerably. Sections 11.1 through 114 present the fundamental theory of linear systems. Much of this material was found in Section 7.6in the second edition, and some additional results are also included here. Sections {1.5 through 11.7 now ‘present the basic theory of the single nth-order equation, making considerable use ‘of the material of the preceding sections. Section 11 8 introduces second-order self-adjoint equations and proceeds through the fundamentals of classical Sturm ‘Theory. We believe that the finear theory is now presented more coherently than in the previous edition 10, An appendix presemts, without proof, he fundamentals of second and third order determinants, ‘The book can be used as a text in several different types of courses. The more or less traditional one-semester introductory course could be based on Chapter ? through. Section 7.4 of Chapter 7if elementary applications are to be included. An alternative fone-semester version omitting applications but including numerical methods and ‘Laplace transforms could be based on Chapters 1,2, 4,6, 7,8, and 9. An introductory ‘course designed to lead quickly to the methods of partial diflerentisl equations could be based on Chapters 1, 2(in part), 4,6, 12, and 14. “The book can also be used as a text in various intermediate courses for juniors and seniors who have already had a one-semesier introduction to the subject. An interme- diate course emphasizing further methods could be based on Chapters 8, 9, 12, 13, and 14. An intermediate course designed as an introduction to fundamental theory could be based on Chapters 10 through 14. We also note that Chapters 13 and 14 can be interchanged advantageously. T am grateful to several anonymous reviewers who made useful comments and suggestions. I thank my cotleagues Wiliam Bonnice and Robert O. Kimball for helpful advies. [also thank my son, Shepley L. Ross, Il, graduate student in mathematics, University of Rochester, Rochester, New York, for his careful reviewing and helpful suggestions, ‘Tam grateful to Solange Abbott for her excellent typing, {am pleased to record my appreciation to Editor Gary Ostedt and the Wiley staff for their constant helpfulness and cooperation. ‘As on several previous occasions, the most thanks goes to my wife who offered encouragement, understanding, patience, and help in many different ways. Thanks, Gin, Shepley L. RossPART ONE FUNDAMENTAL METHODS AND APPLICATIONS ‘One Differential Equations and Their Solutions LL Clasication of Diferential Equations; The Origin and Application LZ Solutions 1.3 Initia-Value Problems, Boundary-Valve Problems, and Existence of Solutions ‘Two First-Order Equations for Which Exact Solutions Are Obtainable 21 Exact Differential Equations and Integrating Factors 22. Separable Equations and Equations Kedveibie Ths Form ~ 23 Linear Equation and Bemoull Equations 24 Special Imegrating Fators and Transformations Three Applications of First-Order Equations 4.1. Orthogonal and Oblique Trajectories 32 Problemsin Mechanics 433. Rate Problems licit Methods of Solving Higher-Order Linear ial Equations AL Basic Theory of Linear Difeential Equacions 42. The Homogeneous Linear Equation with Constant Coeficiente —— CONTENTS —— 102 102 1543. "The Method of Undetermined Coefficients 44 Variation of Parameter: 43. The Cauchy-Euler Equation 46 Statements and Proofs of Theorems on the Second-Order Homogeneous Linear Equation ° Five Applications of Second-Order Linear Differential Equations, with Constant Coefficients 5.1 The Differential Equation ofthe Vibrations of @ Mau on a Spring 52. Free, Undamped Motion 53. Free, Damped Motion 54 Forced Motion 53. Resonance Phenomena 5.6 Blectric Cirewie Problems Six Series Solutions of Linear Differential Equations, 6.1 Power Series Solutions About an Ordinary Point 6.2 Solutions About Singular Points; The Method af Feobenivt 63 Besse’ Equation and Bessel Functions Seven Systems of Linear Differential Equations 2.1 Dilfeentiat Operators and an Operator Method 72 Applications 7.3. Basic Theory of Linear Sysems in Normal Form: Two Equations in Two Unknown Fonction: 74 Homogeneous Linear Systems with Constant Cocfisients: Two Equations in Two Unknown Fanctiont 23. Matrices and Vectors 16 The Matri Method for Homogeneous Linear Sytems with Constant Coeficiente: ‘Two Equations in Two Unknown Functions The Matri Method for Homogencous Linear Systems with Constant Coeficints: ‘Equations in e Unknown Functions Eight Approximate Methods of Solving First-Order Equations 8.1. Graphical Methods 82 Power Series Methods 83. The Medhod of Succesive Approximations 84 Numerical Metheds ‘Nine The Laplace Transform 9.1 Definition, Existence, and Basie Properties ofthe Laplace Transform 92. The Invere Transform ant the Convolution 93 Laplace ‘Transform Selition of Linear Differential Equations with Constant Cocticient 94 Laplace Transform Solution of Linear Systems 179 19 132 199 au. a 2 23 232 264 264 28 290 sot a2 346 355 377 37 se 390 394 an su a1 a 453| cones PART TWO FUNDAMENTAL THEORY AND FURTHER METHODS Ten Existence and Uniqueness Theory 461 | 101 Some Conceps fiom Real Funetion Theory 461 } 102 ‘The Fundamental Bastence and Uniqucness Theorem 473 } 105 Dependence of Solutions on Inti Conditions and an the Funeti 8 104 Extence and Uniguenes Theorems fo Systems and Higher-Order Equations 498 Eleven The Theory of Linear Differential Equations 305 HA Touroduetion 50s 12. Basie Theory ofthe Homogencous Linear Sytem 510 13. Further Theory of the Homogencous Linear System m2 14 The Nonkomogenecus Lineae System 53, 1.3. aie Theory ofthe wh-Order Homogeneous Linea Differasial Equation 48 116 Further Poperis ofthe wh-Order Homogencous Lica Diferetial Equation 558 117. "The mh-Order Nenhomogencous Linear Equation 569 1B. Sturm Theory 373 Twelve Sturm-Liowville Boundary-Value Problems and Fourier Series 588 i 1241, Sturm-Liovle Problems ses j 122. Orthogonal of Characters Factions on ‘ 123 The Bepason oa Fenton na Ser of Ontonormal Function oor { 124 “Tgpeemerc Fourier Sees 8 { Thirteen Nonlinear Dierntial Equations on \ 182. Phase Plane Pats and ii! Poi see | 182 Gaal Pons and Phe Liner Systeme bea 153. Grea Pom and Fao Nonna Sens ost 134 Limit Gees anl Peo Satan esa 133 The Meth o Key a Boga bt Pa Fourteen Partial Differential Equations 15 V4 Some Base Concepts and Examples ns 162. Treated of Sparen of Vrs m 143. Canonical Foo Second-Order Linear quai with Constant Coefiens 148 THA hn Vale bien; Chgracecs : Bt Appendices m Answers a ‘Suggested Reading 807 Index 803———PART ONE——— FUNDAMENTAL METHODS AND APPLICATIONS—— CHAPTER ONE—— Differential Equations and Their Solutions ‘The subject of diflerential equations constitutes a large and very important branch of ‘modern mathematics, From the early days of the calculus the subject has been an area of great theoretical esearch and practical applications, and it continues to beso in our day. This much stated, several questions naturally arise. Just what is a diferent equation and what does it signify? Where and how do differential equations originate and of what use are they? Confronted with a differential equation, what does one do il, how does one do it, and what are the results of such activity? These questions indicate three major aspects of the subject: theory, method, and application. The Purpose ofthis chapter is to introduce the reader tothe basic aspects ofthe subject and. atthe same time give a brief survey of the three aspects just mentioned. In the course of, ‘the chapter, we shall find answers to the general questions raised above, answers that ‘will become more and more meaningful as we proceed with the study of diferential ‘equations in the following chapters. 1.1 CLASSIFICATION OF DIFFERENTIAL EQUATIONS; THEIR ORIGIN AND ‘APPLICATION A. Differential Equations and Their Classi DEFINITION Anequation inzolving derivatives of one or more dependent variables with respect to one or ‘mare independent variables is called a differential equation.* in connection with is bac defiaon, we do ror nclade inthe class of dori equations those tsqustons that ate actly tive tdontes For example, wa excude sich expresons a, Lier = set, Law = ober ad 50 forts * os roar 34 DIFFERENTIAL EQUATIONS AND THEE SOLUTIONS D Example 1.1 For examples of differential equations we lst the following #4 (2) S+n(Z) on dts dix 1 4 58 1 se msint a a3) Pu Ou Ou ee tg a From the bref lis of diflerential equations in Example 1.1 itis clear that the various variables and derivatives involved in a diflerential equation can occur in a variety of ways, Clearly some kind of classification must be made. To begin with, we classify example 1.2 Equations (1.1) and (1.2) aré ordinary differential equations, In Equation (I.1) the variable xis the single independent variable, and yisa dependent variable. In Equation (1.2) the independent variable is ¢, whereas x is dependent DEFINITION A differential equation inoolving partial derivatives of one or more dependent variables ‘with respect to more than one independent variable is called a partial differential equation. > Example 1.3 Equations (1.3) and (1.4) are partial differential equations. In Equation (1.3) the variables sand tare independent variables and visa dependent variable. In Equation (1.4) there are three independent variables: x, y, and 2; in this equation wis dependent. We further classify differs ial equations, both ordinary and partial, according to the ‘order of the highest derivative appearing in the equation. For this purpose we give the following definition1.1 CLASSIFICATION OF DHFERENTIAL EQUATIONS; THEI ORIGIN AND ARUCATION 5 DEFINITION ‘The order of the highest ordered derivative involved ina differential equation iscalled che order of the diferental equation D Example 1.4 ‘The ordinary differential equation (1.1) is of the second order, since the highest derivative involved is a second derivative. Equation (1.2) is an ordinary differential equation of the fourth order. The partial diferentia equations 1.3) and{(1.4)areof the first and second orders, respectively. Proceeding with our study of ordinary differential equations, we now introduce the important concept of linearity applied to such equations. This concept will enable us 0 classify these equations still further. DEFINITION A Yinear ordinary differential equation of order n, in the dependent variable y and the independent variable x, is an equation that is in, or can be expressed in, the form anty de ay 7 y aa) + UD act +7 OyasCaD Gy + aly = Bla, where ag isnot identically zero, Observe (1) that the dependent variable yand its various derivatives occur to the first degree only, (2) that no products of y and/or any of its derivatives are present, and (3) that no transcendental functions of y and/or its derivatives occur. D> Example 1.5 ‘The following ordinary differential equations are both linear. In each case y is the dependent variable. Observe that y and its various derivatives occur tothe first degree only and that no products of y and/or any of its des fy 6564 bye BrtsZsoy-o, as ay Py odd oe Gate grt Zax a) DEFINITION ‘nonlinear ordinary dierent equation san rdlnar diferente equation hats not6 DirremenTiAt EQUATIONS AND THEIR SOLUTIONS. > Example 1.6 ‘The following ordinary diferential equations are all nonlinear: ay a an ey Pres as) a a | Equation (17) i nonlinear because the dependent variable y appears to the second degree in the term 6y", Equation (1.8) owes its nonlinearity to the presence of the term Sidy/dx)?, which involves the third power of the First derivative. Finally, Equation (1.9) is nonlinear because of the term Sy(dy/dx), which involves the product of the dependent variable and its first derivative Linear ordinary differential equations are further classified according to the nature ‘of the coefficients of the dependent variables and their derivatives. For example, Equation ({.5)issaid to be linear with constant coefficients, while Equation (1.6)is linear with variable coeficients B. Origin and Application of Differential Equations Having classified diferential equations in various ways, let us now consider briefly where, and how, such equations actually originate. In this way we shall obtain some indication of the great variety of subjects to which the theory and methods of differential equations may be applied Differential equations occur in connection with numerous problems that are ‘encountered inthe various branches of science and engineering. We indicate a few such problems in the following list, which could easily be extended to fill many pages. rocket, sat oF planet. 1, “The problem of determining the motion of a projecti 2. The problem of determining the charge or current in an electric circuit. 3. The problem of the conduction of heat in & rod or in a slab. 44. The problem of determining the vibrations of a wire or a membrane. 5. The study of the rate of decomposition of a radioactive substance or the rate of growth of a population. 6 The study of the reactions of chemicals. 7. The problem of the determination of curves that have certain geometrical properties. ‘The mathematical formulation of such problems give rise to differential equations. But ust how does this occur? In the situations under consideration ineach of the above problems the objects involved obey certain scientific laws. These laws involve various rates of change of one or more quantities with respect to other quantities. Let use12 sownons 7 call that such rates of change are expressed mathematically by derivatives. In the ‘mathematical formulation of each of the above situations, the various rates of change are thus expressed by various derivatives and the scientific laws themselves become ‘mathematical equations involving derivatives, tha is, differential equations. In this process of mathematical formulation, certain simplifying assumptions generally have to be made in order that the resulting differential equations be tractable. For example, ifthe actual situation ina certain aspect of the problem is ofa relatively complicated nature, we are often forced to modify this by assuming instead an approximate situation that is of a comparatively simple nature. Indeed, certain relatively unimportant aspects of the problem must often be entirely eliminated. The result of such changes from the actual nature of things means that the resulting diflerential equation is actually that of an idealized situation. Nonetheless, the information obtained from such an equation i of the gfeatest value to the scientist A nnatural question now isthe following: How does one obtain useful information froma differential equation? The answer is essentially that itis possible todo s0, one solves the differential equation to obtain a solution; if this is not possible, one uses the theory of differential equations to obtain information about the solution. To understand the meaning ofthis answer, we must discuss what is meant by a solution of 4a diflerential equation; this is done in the next section. Exercises differential Cassify each of the following differential equations as ordinary or parti tqustiossatethe orderofeachequatoncand deerme wheter heaton oder Smideration ea oc nonin 1 La ety ere. 2 Sg syesinx Boo 4 dye yarn 2 (Eee 0 (t= fn 1.2 SOLUTIONS A. Nature of Solutions We now consider the concept of a solution of the nth-order ordinary differential ‘equation,8 owreteniat eQuaTiONs AND THUR SOLUTIONS DEFINITION fondo ft =n a dy ty where Fis areal function ofits (n+ 2) arguments x, 1% oo 1. Let f be a reat function defined forall x ina real inereal I and hasing an nth derivative (and hence aso all lower ordered derivatives) forall x€ 1. The function fis called an explicit solution ofthe differential equation(J.10)on 1if i fulfill the following ‘v0 requirements: FL SOM Loe LCT “ is defined for all x € 1, and FL £9, S'.-- SE] =O 8 Jor all x 1. That is, the substitution of f(x) and its various derivations for y and ts corresponding derivatives respectcely, in (I-10) reduces (10) to an identity on 2. relation olx,») = Os caled an implicit solution of 10) this relation defies cat least one eal function f ofthe variable x on an interoal sack hat this function i an explicit solution of (1.10) on this interval 3. Botkexplcit solutions and implicit solutions will usually be called simply solutions Roughly speaking, then, we may say that a solution ofthe differential equation (1-10) isa relation explicit o implicit—between x and y, not containing derivatives, which identically satisfies (1.10) D Example 1.7 “The function f defined for all real x by fle) = 2sin x + 3008 x aan is an explicit solution ofthe differential equation ay Shtys0 ary fora ral x. First note that is defined and has.a second derivative forall real x. Next ‘observe that J") = 208 x —3 sin x, Sta) = ~2sin x— 3.08 x. ‘Upon substituting f(x) ford? y/dx? and f(x) for yin the diferental equation (1-12), it reduces to the identity (~2sin x ~ 3.08 x) + (2 sin x-+ 3.08 x) = 0,12 sorutions 9 ‘which holds for all real x. Thus the function f defined by (1.11) isan explicit solution of the differential equation (1.12) forall real x. > Example 1.8 ‘The relation x+y? 2550 13) is an implicit solution of the differential equation ay x4 yZn0 a9 ‘on the interval I defined by —5 Example 1.17 Problem. Find a solution f of the diferemtial equation ay Ram (2y such that at x = I this solution f has the value 4, Explanation, First let us be certain that we thoroughly understand this problem. We seck a real function f which fulills the two following requirements: 1. Thefonetion f must satisfy the differential equation (1.21) Thats, the function f must be such that J") = 2x fr ll real x ina ral interval 2. Thefunction f must havethe value 4atx = 1. Thatis the function f must besuch that £0) = 4 Notation. The stated problem may be expressed in the following somewhat abbreviated notation: ay Bory, ax yO) = In this notation we may regard y as representing the desired solution. Thea the differential equation itself obviously represents requirement 1, and the statement (l= 4 stands for requirement 2. More specifically, the notation y(1) = 4 states that the desired solution y must have the value 4 at x = 1; that is, y= 4 at x Solution. We observed in Example 1.9 that the difleential equation (1.21) has a ‘one-parameter family of solutions which we write as yextte, (22) where cisan arbitrary constant, and that each ofthese solutions satisfies requirement L Let us now attempt to determine the constant ¢ so that (1,22) satisfies requirement 2, thats, y = 4.at x= 1, Substituting x = 1, y = 4 into (1.22), we obtain 4 = | + ¢,and hhence ¢ = 3, Now substituting the value c= 3 thus determined back into (1.22), we obtain yan,16 { } DIFFERENTIAL EQUATIONS AND THER SOLUTIONS which is indeed a solution ofthe differential equation (1.21), which has the value 4 at In other Words, the function f defined by Soy +3, satisfies both of the requirements set forth in the problem. Comment on Requirement 2 and its Notation. Ina problem of this type, require- ment 2 is regarded as a supplementary condition that the solution of the diferential equation must also satisy. The abbreviated notation y(1) = 4, which we used to express this condition, is in some way undesirable, but ithas the advantages of being both customary and convenient. In the application of both first-and higher-order differential equations the problems most frequently encountered are similar to the above introductory problemin that they involve hoth a differential equation and one or more supplementary conditions which the solution of the given differential equation must satisfy. 1C all of the associated ‘supplementary conditions relate to one x value, the problem is called an intial-zalue ‘rable (or one-point boundary-valve problem). If the conditions relate to ro dif- ferent x values, the problem is called a two-point boundary-value problem (or simply a boundary-valve problem). We shall illustrate these concepts with examples and then ‘consider one such type of problem in detail. Concerning notation, we generally employ abbreviated notations for the supplementary conditions that are similar to the ab- breviated notation introduced in Example 1.11 > Example 1.12 0) p= -4 This problem consists in finding a solution of the differential equation Which assumes the value 3at x = | and whose first derivative assumes the value —4 at Both of these conditions relate to one x value, namely, x = I. Thus this is an initial-value problem, We shall see later that this problem has @ unique solution. Example 1.1313 IMTIALVALUE Prom, soUNDARY.vALUE PROBLEMS 17 In this problem we again seck a solution of the same differential equation, but this time the solution must assume the value 1 at x =O and the value $ at x = n/2. That is, the conditions relate to the two different x values, 0 and 1/2. This is a (two-point) boundary-value problem. This problem also has a unique solution; but the boundary value problem i: Paya, YO=I yln)= 5, has no solution at all! This simple fact may lead one to the correct conclusion that boundary-value problems are not to be taken lightly! We now turn toa mote detailed consideration of the initial-value problem fora frst- order dilferential equation. DEFINITION Consider the first-order differential equation és =f (123) where {is «continuous function of x and y in some domain® D ofthe xy plane: and let (50, 3) be a point of D. The initial-value problem associated with (7.23) is to find ‘solution ofthe differential equation (1.23) defined on some real interval containing Xp, ‘and satisfying the inital condition $600) = Yo! In the customary abbreviated notation, this inital-ualue problem may be written ‘ eo Seah Wx) = Yo. ‘osolvethis problem, wemust inda fonction that not onl satsfos the diferentiat equation (1.23) but that also satsfes the initial condition that it haste value y when x has value x. The gsometrc interpretation of th initial condition, and hence ofthe centr initalvalye problem, i casly understood. The graph of the dsied solution must pas through the pont with coordinates (x, ye). That i, interpreted geometi- cally, the initial-value problem st find an integrl curve ofthe diferntial equation (1.28) that passes theough the point (x) “Themethod of actually finding the desired solution depends upon thenatue of the differential equation ofthe problem, that i, upon the form off}, Cetain special types of iflerential equations have a one-parameter family of solutions whose equation may be found exactly by following definite procedures (ee Chapter 2). the Sifferetial equation ofthe problemi of some such special type, one st obiain the equation of ts one-parameter family of solutions and then applies the initial condition "Nema an open, Connected at For ove unarliar wih ach concept, O may be eed the interior of Some simple closed cure inthe plore.18 wrenentia. tQuaTions AND THEIR SOLUTIONS {othisequationin an attempt to obtaina“particular” solution @ that satisfies the entire initial-value problem. We shall explain this situation more precisely in the next paragraph. Before doing so, however, we point out that in general one cannot find ‘he equation of a one-parameter family of solutions of the differential equation; approximate methods must then be used (see Chapter 8) Now suppose one can determine the equation als, no) = 0 2) cof a one-parameter family of solutions of the differential equation of the problem. ‘Then, since the initial condition requires that y= yp at x= xo, we let x = Xp and 13 Vo in(1.24) and thereby obtain Bo.¥od = Solving this for cin general we obtain a pasticular value of ¢ which we denote here by x, Wenow replace the arbitrary constant c by the particular constant cy in (1.24, thus ‘obtaining the particular solution 4s, Yea) = 0. ‘The particular explicit solution satistying the two conditions (differential equation and {niial condition) of the problem is then determined from this, if possible. We have already solved one initial-value problem in Example 1.11. We now give another example in order to illustrate the concepts and provedures more thoroughly > Example 1.14 Solve the initial-value problem (125) (126) given that the differential equation(1.25)has a one-parameter family of solutions which -may be written in the form, (127) ‘The condition (1.26) means that we seek the solution of (.25)such that y = 4at x = 3 ‘Thus the pair of values (3,4) must satisy the relation (1.27) Substituting x = 3 and 4 into (1.27), we find vay 9+ 16=c? or c ‘Now substituting this value of c? into (1.27), we have xt yha 2s Solving this for y, we obtain yas VB—H ‘Obviously the positive sign must be chosen to give y the value +4 at x = 3. Thus the function f defined by fl) = JB ~Sex Example 1.15 Con inte den A alte Bovey, wl Let us apply Theorem 11. We first check the hypothesis. Here f(x, y) = x? + y* sn Ty oh of ton and fy ae ots yon of the xy plane. The initial condition y(1) = 3 means that xo = Land yp = 3,and the point (1, 3) certainly lies in some such domain D. Thus all hypotheses are satisfied and the conclusion holds. That is, there is a unique solution @ of the diferential equation dyjdx = x* + y?, defined on some interval |x — 1] Example 1.16 Consider the tio problems: ayy L Rage nr yoy 2 Bete rome Here fx Sewn zin ane ‘These functions are both continuous except for x = 0 (that is, along the y axis). In problem 1.x = 1, yo = 2. The square of side I centered about (1,2) does not contain the y axis, and so both f and df/éy satisty the required hypotheses inthis square. Is interior may thus be taken to be the domain D of Theorem 11; and (1,2) certainly les within it, Thus the conclusion of Theorem 1.1 applies to problem 1 and we know the problem has @ unique solution defined in some sufficiently small interval about xp = J Now let us turn to problem 2. Here xy = 0, yo = 2.At this point neither f nor dff2y are continuous. In other words, the point (0,2) cannot be included in a domain D where ‘the required hypotheses are satisfied. Thus we can no: conclude from Theorem 1.1 that22 overex, equations ANO THEHE SOLUTIONS problem 2 has a solution, We are not saying that it does not have one. Theorem 1.1 simply gives no information one way or the other. Exercises 1. Show that ya det +20 yo)= 6, YO) =2. Is y = 2% + 4e-* also a solution of this problem? Explain why or why not 2. Given that every solution of uae Rt ya2 may be written in the form y = (x? 4 de~*, for same choice of the arbi constant c solve the following initial-value problems: ay oF ® yee @ Pa yeres, ty Pyare, 0) = 2 wale +3. 4, Given that every solution af ey ay _ oy 8 yao ‘may be written in the form pecs tee, for some choice of the arbitrary constants c, and 3, solve the following initial- value problems: ®y @ Py _ay & -Z- nyo 10) yOr6 4. Every solution ofthe diferential equation @y Gitrno( Reece Be esc ee OSE aEnREEE 13 INITAL-VALUE PROBLEMS, ROUNDARY-VALUE Poss 23 may be written in the form y = ¢, sinx + ¢; os, for some choice of the arbitrary ‘constants c, and c. Using this information, show that boundary problems (a) and. (©) possess solutions bur that (c) does-n0:. a © Baye yO) = 0, yO) = 1, Wen/2) 41 YR) = “ © Bayeo 30) =0, 0) 1 “ Given that every solution of Py yi dy gt Eh - 3 Sh 4 6% oye ‘may be writtenintheform y = ¢,x + ¢,x7 + ex? forsome choice of the arbitrary constan's ¢,,¢3, and ¢s, solve the initial-value problem consisting of the above dlifferential equation plus the three conditions yQ)=0, y@=2 y"'A=6 . Apply Theorem 1.1 to show that each of the following initial-value problems has 2 unique solution defined on some suficently small inverval |x ~ 1| 0. (8) Carefully graph the solution for which ¢ = 0. Then, using this particular staph, also graph the solutions for which ¢ = 1,¢ = 2, and ¢ = 3.=== CHAPTER TWO—— First-Order Equations for Which Exact Solutions Are Obtainable In this chapter we consider certain basic types of first-order equations for which exact solutions may be obtained by definite procedures. The purpose ofthis chapteristo gain ability to recognize these various types and to apply the corresponding methods of solutions. OF the types considered here, the so-called exact equations considered in Section 2.1 arcin asense the most basic, while the separable equations of Section 2.2are {na sense the “easiest.” The most important, from the point of view of applications, are the separable equations of Section 2.2 and the linear equations of Section 23. The remaining types are of various very special forms, and the correponding methods of solution involve various devices. In short, we might describe this chapter as a col- lection of special “methods,” “devices,” “tricks,” oF “recipes. in descending order of kindness! EXACT DIFFERENTIAL EQUATIONS AND INTEGRATING FACTORS, A, Standard Forms of Fitst-Order Differential Equations The first-order differential equations to be studied in this chapter may be expressed in cither the derivative form Pesan ey or he iteremia form Mos hax + NO Nd =O. an 2526 masronoer EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE ‘An equation in one of these forms may readily be written in the other form. For example, the equation ay ety & is ofthe form (2.1). It may be written (e+ y)ax+ aay which i of the form (22). The equation (sin + yds + (4 39) y =0, which i of the frm (22), may be written in the form (21) a8, In the form (2.1) itis clear from the notation itself that is regarded as the dependent variable and x as the independent one; but in the form (2.2) we may actually regard cither variable as the dependent one and the other as the independent. However, in this text, in all differential equations of the form (2.2) in x and y, we shall regard y as dependent and x as independent, unless the contrary is specifically tated B. Exact Differential Equations DEFINITION Let F be a function of two real variables such that F has cominuous first partial derivatives in a domain D. The total dillerental dF of the function F is defined by the formula 9 = ED an 4 FLED cs {for ait(x, )¢ D. > ixampe 21 ~ Let F be the function of two teal variables defined by Fis, y)= ay? + Bey for al real (x, 3). Then BFC sg gery, SFE : is 675, SRD e aey 4 20% ‘and the total differential dF is defined by F(x, 9) = (9? + 6x7y) do + Oxy + 2x) dy for all real (x)| | I | } { i | | i | 2.4 WACK DIFFERENTIAL EQUATIONS AND INTEGRATING FACTORS 27 DEFINITION The expression M(x, y) dx + NOs y) dy @3) is called an exact differential ina domain D if there exists a function F of two real vari- ‘ables such that this expression equals the total differential dF(x,y) for all (x,y) €D. That is, expression (2.3) is an exact differential in D if there exists @ function F such that 2PM es.) and FED 9 aces fue yal a cg cated ope aa age ce M(x, y)dx + N(x, ypdy = 0 seated ana trent eution > Example 2.2 The differential equation yi dx + 2xpdy=0 es is an exact diferential equation, since the expression y? dx + 2xy dy is an exact Example 23 We apply the exactness criterion (2.7) to Equations (2.4) and (2.5), introduced in Example 22. For the equation y? dx + 2xydy=0 (24) we have Mey) N(s,9) = 29, OMG 9) _ 9, _ ANU 9) ey a for all(x,»). Thus Equation (24) is exact in every rectangular domain D. On the other hand, for the equa pdx + 2xdy=0, as we have Mex y= NOs 9) = 2x aMix 9) antx,») a jon (2.8) isnot exact in any rectangular domain D, 142 for all(x, 9). Thus Equi12.1. BRACT DIFERENTIAL EQUATIONS ANO WwTEGaATING FACTORS 3 > example 24 Consider the differential equation Qxsin y+ ye") de + (x? cosy + 3y%e dy = 0. Here M(x y)= 2x sin y + ye N(x y)= x? 60s y+ 39%e, ames 9) ents.) ay a inevey rectangular domain D Ths his diferent equation is exc in vty such = 2x cos y + 3yte* = ‘These examples illustrate the use ofthe test given by (2-7)for determining whether or not anequation of the form M(x, y) dx + N(x, y) dy = Oisexact.Itshould be observed ‘that the equation must be in the standard form M(x, y) dx + N(x, y) dy = Oin order to Use the exactness test (27). Note this carefully: an equation may be encountered inthe nonstandard form M(x, ») dx = N(x, »)dy, and in this form the test (2.7) does not apply. C. The Solution of Exact Differential Equations ‘Now that we havea test with which to determine exactness let us proceed to solve exact differential equations. If the equation M(x, ypde + N(x, »)dy =O is exact in a reétangular domain D, then there exists a fonction F such that FFL») _ sayy oF») i SR Messy and 7 « Nix 9) forall ye. ‘Then the equation may be written BOD gy POD yay oxsingy dF 91= ae ae + y dy=0 orsimply dF (x, y) = 0. ‘The celation F(x, #} = cis obviously a solution of this, wherecis an arbitrary constant ‘We summarize this observation in the following theorem. ‘THEOREM 2.2 ‘Suppose the differential equation M(x, y) dx + NIx, y) dy = 0 satisfies the differentia. Dilley requirements of Theorem 2.1 and s exact wn a rectangular domain D. Then a one- parameter family of solutions of this differential equation is gioen by F(x, ») = ¢, where F isa function such that aF i, Ft FEDS Mex, 9 and : 2D Nix,y) for ail (x ye. ‘and ets an arbitrary constant.Fns1-0806R EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE Referring to Theorem 2.1, we observe that F(x, y) is given by formula (2.13) However, in solving exact differential equations its neither necessary nar desirable to Use this formule. Instead one obtains F(x, y) either by proceeding asin the proof of Theorem 2.1, Part 2,orby the so-called “method of grouping,” which willbe explained inthe following examples > Example 25 Solve the equation Bx? + dy) dx + Ox? + Iy)dy = 0. ‘Our first éuty is to determine whether or not the equation is exact, Here M(x, y) = 3x? + 4p, NOx 9) = 28? +29, Mix, 9) ant 9) a x for all real (x, yj and so the equation is exact in every rectangular domain D. Thus We must find F such that 66,9) oF») ey = Misys) = 38! + 44y- and = Nu yb = 208 429 From the fist of these, Fo = futon ax + 00)= foe 4+ 4ny) dx + 00) Bey FOOL, 259) _ aya , HOU? eur dy ‘Buc we must have Thus ‘Thus #(y) = ¥? + co, where cp is an arbitrary constant, and so : Fix, ax + Bet ty? +00ws AND intecRATING Factors 33. 24 Exact oUrimeNALEQUAN Hence a one-parameter family of solution is F(x, y) a4 ety + yb eg me} Combining the constants ¢y and cy we may write this solution as ety ts Pac, wheree = in generality by taking cy = 0 and writing $y) procedure, ois anarbitrary constant. Thestudent will observe that there sno oss y3, We now consider an alternative Method of Grouping. We shall now solve the differential equation of this exam ple by grouping the terms in such a way that its let member appears as the sum of certain exact differentials. We write the diferential equation Gxt + dxyl de + Ox? 4 2y)dy = 0 in the form 3x? dx + (dy dx + 2x? dy) + 2y dv = 0. We now recognize this as dix?) + d(2x?y) + diy?) where ¢ is an arbitrary constant, or dls? + Ixty + y?) = dle. a, From this we have at once xtadty tyme, Clearly this procedure is much quicker, but it requires a good “working knowledge” of differentials and a certain amount of ingenuity to determine just how the terms should be grouped. The standard method may require more "work and take longer, but its perfectly straightforward. [tis recommended for those who like to follow a pattern and for those who have a tendency to jump at conculsions. Just to make certain that we have both procedures well n hand, we shall consider an initial-value problem involving an exact differential equation, > Example 2.6 Solve the initial-value problem (De cos y+ Ixty) dx + (x? — x3 sin y= yhdy = 0, 0) = 2 ‘We fist observe thatthe equation is exact in every rectangular domain D, since aM») iny + 3x2 = ON) Spe = anny 4 act = Rs forall rel (x, 9)34. emst-onote EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE Standard Method. We must find F such that Fon 9) a) a My) = 2ec08 y+ 3x4 and ree (ea ee Bo Mag) =~ x7sin yp Then Fe,y)= f Mix, ix +6) = fexcosy +349 2x4 649 = eosy4yt 60), 2FO ys >, dal) ssiny tt OU But also FOI) _ Nyy yy ex =x gin FED Ns yaw yoy and so and hence oun B00, Thus Fes, y) = x eos y +29 2 + oo Hence a one-paraineter family of solutions is Fix, 9) = ¢,, which may be expressed as sPeosy + ty Applying the initial condition y = 2 when x ~ 0, we find ¢ = —2 Thus the solution of ‘he given initial-value problem is sPoos yey Method of Grouping, We group the terms as follows: ede = xtsin yds) + Bx4y dy #8 dy vey21 PRACT DUFERENTIAL EQUATIONS AND INTEGRATING FACTORS 35. ‘Thus we have atx? os ») + ax*y) — (5) = dos and so ew yeey ee sx one-parameter family of solutions of the differential equation. Of course the intial condition y(0) = 2 again yields the particular solution already obtained D. Integrating Factors Given the differential equation M(x 3) dx + Nex y)dy =0, aNtxy) M(x. 9) v then the equation is exact and we can obtain a one-parameter family of solutions by fone of the procedures explained above, But if M(x, 9), ONG 39) 4 ONE ey ex then the equation is nor exact and the above procedures do not apply. What shall we do ‘in such a case? Perhaps we can multiply (he nonexact equation by some expression that will ransform it into an essentially equivalent exact equation. If so, we can proceed to * solve the resulting exact equation by one of the above procedures. Let us consider ‘again the equation yax + 2xdy =0, es which was introduced in Example 2.2. In that example we observed that thisequation is not exact. However, if we multiply Equation (2.5) by y, itis transformed into the ‘essentially equivalent equation vy dx + xy dy =0, e4 hich is exact (see Example 2.2). Since this resulting exact equation (2.4) is integrable, wwe call y an integrating factor of Equation (2.5). In general, we have the following definition: DEFINITION If the diferencia equation Mex. yh dx + Nlxs)dy = 0 1a36 FIRST.ORDUR EQUATIONS FO WHHCH EAACT SOLUTIONS ARE OBTAINABLE is not exact in a domain D but the differential equation HG Mx, 9) dx + ls SINGH WD dy 15) isexact nD, then x, ybis called an integrating factor of the differential equation (2.14). > Example 2.7 Consider the diferent ‘equation Gy + day?) de + (2x 4 3x y dy ‘This equation is ofthe form (2.14), where 2.16) Mix, y) = 3y + 4xy?, N(x, y) = 2x + 3x7y, sie aoa aM by ox 0, Equation (2.16) is not exact in any rectangular exc for (such that 2x9 + omain D. Let u(x, ») x2y, Then the corresponding diflerential equation of the form (2.15) is Gx2y? $ay?) de + Qx°y + ety) dy, ‘This equation is exact in every rectangular domain D, since Ante, YMOx 9) Peau oy ice ay ney? = Mis for al real (x,y), Hence u(x, y) = x?y isan integrating factor of Equation (2.16), Multiplication of @ nonexact diferential equation by an integrating factor thus transforms the nonexact equation into an exact one, We have referred to this result- ing exact equation as “essentially equivalent” to the original. This so-called essentially equivalent exact equation has the same one-parameter family of solutions asthe non- exact original. However, the multiplication of the original equation by the integrating factor may resultin ether (I) the loss of (one oF more) solutions ofthe origina, or (2}the ‘ain of (one or more) functions which are solutions ofthe “new” equation but nar of the Criginal, or (3) both of these phenomena, Hence, whenever we transform a nonexact {equation into an exact one by multiplication by an integrating factor, we should check carefully to determine whether any solutions may have been lost or gained. We shall illustrate an important special case of these phenomena when we consider separable equations in Section 22. See also Exercise 22 atthe end of this section. ‘The question now arises: How isan integrating factor found? We shall not attempt to answer this question at this time. Instead we shall proceed to a study of the important ‘lass of separable equations in Section 2.2and linear equations in Section 2.3, We shall See that separable equations always possess integrating factors that are perleetly{LY ERACT DIFFERENTIAL EQUATIONS AND INTEGRATING Factors 377 ‘obvious, while linear equations always have integrating factors of a certain special orm, We shall return to the question raised above in Section 2.4. Our object here has ‘been merely to introduce the concept of an integrating factor. Exercises In Exercises 1-10 determine whether or not each ofthe given equations is exact; solve those that are exact. Gx + 29)dx4 Qe Ady =O. (07 + 3)de + Oxy — 4) dy =. (xy + Ide +O + ay) dy =O. Gey sad — Wr td (6xy + 29? de + Bx? + dxy - .dy=O. (0+ Neos dr + 20sin rd? = 0. (yse2? x + see xtan x) dx + (tan x +29) dy = 0 G)aGinjond » (eter (SP)enn reat este 10 WL a atts! —yayea Solve he nivale problems in Entei 11-16 1. Qxy-Ddx +7 +4)dy=0, yl =2 12. Gx4y? =p 42a) dx Rey — Say? + Ddy=O, -2)= 13. Qysin xcosx + y*sins)dx + (sin? x—27c0s x) dy m0, y(0) 3 Get b2e + yde He HI HO, O=E aan ene 15, Jee (FSR ayeo nan 20 ett ga 16, Ede BO yn, y=8 = y 17. In each of the following equations determine the constant A such that the equation is exact, and solve the resulting exact equation: fa) (x? + 3xyide + (Ax? + 4y) dy =0, wy (deed)ane(SH)ayu38 {RST-ORDER EQUATIONS FOR WHICH EXACT SOLLIIONS ARE OBTAINABLE 18. In each of the following equations determine the constant A such that the equation is exact, and solve the resulting exact equation: (a) (Axty + 2y*) de + 60° + dry) dy = 0. © (S+h)e+(S-Lo 19, Incach ofthe following equations determine the most general function N(x, ) such that the equation exact: 0. (a) (2 +29") de + NG Way =O. (b) (x#y"F + xy) dx + NG yb dy =O. 20, In each of the following equations determine the most general function M(x, ») such that the equation is exact: (a) Moxy) dx + xy? + xtyhdy (0) M(x, y) dx + Qye" + ye dy 21, Consider the differential equation (4x + 3y?)dx + Day dy =0. (a) Show that this equation is not exact (©) Find an integrating factor of the form x", where n is a postive integer. (€) Multiply the given equation through by the integrating factor found in (6) and solve the resulting exact equation. 22, Consider the differential equation (2 + 2x) de = x? dy =0, (a) Show that this equation is not exact. (b) Multiply the given equation through by y*, where m isan integer, and then determine m so that y" is an integrating factor of the given equation. (6) Multiply the given equation through by the integrating factor found in (b) and solve the resulting exact equation. (4) Show that y = 0s a solution of the original nonexact equation but is not a Solution of the essentially equivalent exact equation found in step (c (¢) Graph several integral curves of the original equation, including all those ‘whose equations are (or can be written) in some “special” form. 23. Consider a differential equation ofthe form Ly xf? + y?)) dx + ofl? + 94) — x) dy = (2). Show that an equation of this form is not exact. (©). Show that 1/(x? + y?)is an integrating factor of an equation ofthis form. 24, Use the result of Exercise 2(b) to solve the equation [yb x08 + 927] de + O96? + 92) ad dy 2.122 SEPARABLE EQUATIONS AND EQUATIONS REDUCIELE TO ms FORM 39. 2.2 SEPARABLE EQUATIONS AND EQUATIONS REDUCIBLE TO THIS FORM ‘A. Separable Equations DEFINITION An equation of the form FO)G(y) dx + Sls) dy = 0 can 7 with variables separable or simply a separable equ is called an equat For example, the equation (x — 4)y4 de — xy? ~ 3) dy = O is a separable equa tion {In general the separable equation (2.17) is not exact, but it possesses an obvious integrating factor, namely 1/f(2)G(y). For if we multiply Equation (217) by this expression, we separate the variables, reducing (2.17) to the essentially equivalent equation FO) te 4 3) ay m0 as) Fey" 6) ‘This equation is exact, since ofa ]-0- 2 [a oy | FO). ax LG), Denoting F(x)/f(x) by M(x) and g(y)/G(y) by N(y) Equation (2.18) takes the form Mats NOj tp = 0 Shee iss funchon ef tony Sods uncon o yon acerca tet Sones way uno fuera [aoa 19 here cis the arbitrary constant. Thus the problem of finding such a family of solutions of the separable equation (2.17) has reduced to that of performing the integrations indicated in Equation 2.19). Its in this sense that separable equations ae the simplest firstorder differential equations. Since we obtained the separated exact equation (218) from the nonexact equation (2.17) by multiplying (2.17) by the integrating factor 1/(x)G4)), solutions may have ‘been Jost or gained in this process. We now consider this more carefully. In formally ‘multiplying by the integrating factor 1/(s)G(y), we actually divided by /(2)G(y). We did this under the tacit assumption that neither /(x) nor G(y) is zero; and, under this assumption, we proceeded to obtain the one-parameter family of solutions given by (2.19), Now, we should investigate the possible loss or gain of solutions that may have ‘curred in this formal process. In particular, regarding yas the dependent variable as usual, we consider the situation that occurs if Gt) is zero. Writing the original dif- ferential equation 2.17) in the derivative form Flag a + FIG) = 0,40 "RST-ORDER EQUATIONS FOR WHICH EXACT SOLUTIONS ARE DBTANABLE we immediately note the following: If yy is any real number such that G(y) = 0, then = Yo i a (constant solution of the original differential equation; and this solution ‘may (or may not) have been lost in the formal separation process. Im finding a one-parameter family of solutions a separable equation, we shall always make the assumption that any factors by which we divide in the formal ‘separation process are not zero. Then we must find the solutions y = yp of the equation G(s) = 0 and determine whether any of these are solutions of the original equation which were lost inthe formal separation process. D Example 2.8, Solve the equation (= dy dx? dy =0. ‘The equation is separable; separating the variables by dividing by x*y*, we obtain Fax dx O77 = 3 dy = 0. Integrating, we have the one-parameter family of solutions where isthe arbitrary constant. In dividing by x?y* in the separation process, we assumed that x° ¥ O and y* # 0, Wenow consider the solution y = Oof y* = 0. Itisnota member of the one-parameter family of solutions which we obtained, However, writing the original diferential equation of the problem in the derivative form dy _ @—4y* & SG? ‘obvious that y = 0 sa solution ofthe original equation. We concifde that itis 2 solution which was lost in the separation process. > Example 2.9 Solve the initial-value problem that consists ofthe differential equation sin yde + (x4 + 1) cos ydy =0 (229) and the initial condition vt) =5. ean We first obtain a one-parameter family of solutions ofthe differential equation (2.20),212. SEPARABLE EQUATIONS AND FQUANONS sEDLCBLE TO THs roms 41 Separating the variables by dividing by (x? + 1) sin y, we obtain x cory ayn de 4 EP ay m0, Fatt Say? Thus xéx | (oor ys, Pa? | siny Om where co is an arbitrary constant, Recall that ft emuiee an win tt Then, carrying out the integrations, we find Fln(x? +1) + Injsin y] =e, (2.22) We could leave the family of solutions inthis form, but we can putt ina neater form in the following way. Since each term of the let member of this equation involves the logatithm of a function, it would seem reasonable that something might be accom- plished by writing the arbitrary constant cy in the form ln |c,|. This we do, obtaining Hints? +1) + In [sin ») = tn Ley) Multiplying by 2, we have In(x? + 1) +2 In|sin y| = 2m fe. Since 24m sin yl = In (sin »)?, and 2imiey| =Incf = Ine, where c=d20, we now have In (x? 41) + Insin? Since In A 4+ In B = In AB, this equation may be written In(x? + sin? y = Ine. From this we have at once (+ Dsin? y Clearly (2.23) is ofa neater form than (2.2) In dividing by (x? + I)sin y in the separation process, we assumed that sin y # 0 Now consider the solvtions of sin y=0, These are given by y= nr (n=O, 21, £2,...), Writing the original differential equation (220) in the derivative'form, iis lear that each ofthese solutions y = n(n = 0, #1, $2,...)of sin y = Disa constant solution of the original differential equation, Now, each of these constant solutions 223)42 | FST-ORDER EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE y= mis a member of the one-parameter family (2.23) of solutions of (20) for ¢ = 0, ‘Thus none of these solutions was lost in the separation process. ‘We now apply the initia! condition (2.2) to the family of solutions (2.23). We have (1 + Hsin? land so.c = 2. Therefore the solution of the initial-value problem under consideration is Ge? + Dsin? y = 2, B. Homogeneous Equations We now consider a class of differential equations that can be reduced to separable ‘equations by a change of variables. DEFINITION The first-order differential equation M(x, 9) dx + N(x, y) dy = 0 ie said to be homo: _peneous when written inthe derivative form (dy dx) = f(s, )) there exist function 9 Such that f(y) an be expressed inthe form g( 3/3). Example 2.10 ‘The differential equation (x* ~ 3y*) dx + 2xy dy = Ois homogencous. To soe this, We first write this equation in the derivative form Now observing that veneneey ESQ 2xy Qe dy 2x) 20, wie se that the diferentil equation under consideration may be written as dy _3/y\_ 11) #-3(:) 2G) in which the right member is of the form g(y/'x) for a certain function g Example 2.11 The equation + SPF F de edy 0 is homogeneous: When writen inthe form ye VPTP am +122. SEPARABLE EQUATIONS AND EQUATIONS REDUCIBLE TO THIS FoRM 43, the right member may be expressed as 14 vEar ve depending on the sign of x. This is obviously of the form g(y/2). Before proceeding to the actual solution of homogeneous equations we shall consider slightly different procedure for recognizing such equations. A function F is called homogeneous of degree nif Fits, 3) = F(x, 3) This means that if tx and ty are substituted for x and 4, respectively, in F(x ), and if ris then factored out, the other factor that remains isthe original expression F(x, 9) tse For example the function F given by F(x, y) = x? + y? is homogeneous of degree 2, since F(x, 19) = (9)? + (0)? = FG? + 97) = PFO). "Now suppose the functions M and NV in the differential equation M(x, y)dx + N(x, )dy = 0 are both homogencous ofthe same degree n. Then since M(¢x, (9) = eMC, 9) if we let t= 1s, we hive 1 6} Mex) u(bsbs) ule and from this we at once obtain ses () mld nesre(’)"n(u2) Now writing the differential equation M(x, y) dx + N(x») dy = On the form ay _ Mts») a” NEY)" Likewise, we find we find re Clearly the expression on the right is of the form g(/), and so the equation M(x, ») dx + N(x, y) dy = 0 is homogeneous in the sense of the original definition of homo-44 sinst-onven eQuaTIONS FOR WHUCH EXACT SOLUTIONS ARE OBTAINABLE geneity, Thus we conclude that if M and N in M(x, y)dx + N(x, y}dy = 0 are both homogeneous functions of the same degree n, then the diferential equation is & homogeneous differential equation. Let us now look back at Examples 2.10 and 2.11 in this light. In Example 2.10, M(x, ») = x? ~ 3y? and N(x, ») = 2xy. Both M and N are homogeneous of degree 2. ‘Thus we know at once that the equation (x ~ 3y7) dx + 2xy dy = Ois a homogeneous equation. In Example 2.11, M(x, y)= y+ /37 + yFand N(x, 9) = —x. Clearly Nis homogeneous of degree 1. Since Mlix.te) = t+ JOP FOP = dy + fe we see that M is also homogeneous of degree 1. Thus we conclude that the equation (7+ JP + VF) dx = xdy = 0 Mea is indeed homogeneous. We now show that every homogencous equation can be reduced to a separable ‘equation by proving the following theorem. ‘THEOREM 2.3 y M(x, y) de + NG ¥) dy = 0 224 is a homogeneous equation, then the change of variables y = wx transforms (2.24) into a separable equation in the variables v and x, Proof. Since M(x, y) dx + N(x, ») dy = Ois homogeneous, it may be written in the ‘on Let y= ox Then ay att ee and (2.24 becomes ae oat =o fo ~ of0)] dx + x dv = 0. “This equation is separable. Separating the variables we obtain de (225) OED.22, SEPARABLE EQUATIONS AND EQUATIONS REDUCIBLE TO THIS FORM 45; Thus to solve a homogeneous differential equation of the form (2.24), we let p = vx ‘and transform the homogeneous equation into a separable equation of the form (2.25). From this we have de fax : feta JE where cis an arbitrary constant, Letting F(v) denote Sim ‘and returning to the original dependent variable y, the solution takes the form #(2) +9 11c > Example 2.12 Solve the equation (t= 3p) de + Day ‘We have already observed that this equation is homogeneous, Writing it in the form #242 ay and letting y = vx, we obtain or, finally, “This guavon is separable. Separating the variable, we obtain Dodo de Pl x Iotegratng, we fot Info? — 1] = In |x| + Intel, snd hence lo? = = text,[RST ORDUR EQUATIONS FOR WHICH FRACT SOLUTIONS ARE OBTANARLE where cis an arbitrary constant, The reader should observe that no solutions were lost in the separation process. Now, replacing v by y/x we obtain the solutions in the fom be- |y? — x7 |= Leela? If y > x > 0, then this may be expressed somewhat more simply as potent. D Example 2.13 Solve the initial-value problem 4 JF) de — xay = 0, yy =0, ‘We have seen that the differential equation ishomogencous. As before, we write it in the form Since the initial x value is 1, we consider x > Oand take x = J and obtain ay Bats f-Q) Wellet y = ox and obtain Separating variables, we find wae feet x ‘Using tables, we perform the required integrations to obtain Injo+ Jo FI] = In jx} + Ine}, + JPET22. SEPARABLE EQUATIONS AND EQUATIONS aeDUCILE TO THIS FORM 47 Now replacing v by y/x, we obtain the general solution ofthe differential equation in the form yo be Ea yt (OEP = ox? ‘The intial condition requires that y = 0 when x = 1. This gives c = 1,and hence yt VF from which it follows that yoo? Solve each of the differential equations in Exercises 1-14 1 dye tot + Day bey tet yt Dede $ OF 420 dy WO. 2s? + 1dr +64 + Ids = a 3 4. ose yd + seoxdy = 0. 5. tanOdr-+ 2rd0 =0, 6. (e+ Hoos udu + eMlsinu + Ded 2. 8 9. Gh AY + Ne + WO? + Be Day =O. a Oxy +397) dx - Oxy + x4) dy = 0. 10. 9 du (u? — we) do = 0. (stan2s y)ax—ady = 12, Qs 42m + eA) ds 4 (6? + 2st Dd 13. (249 RE Pde — xy SPE dy 14 (Vet yt Jem ax t (lem y~ Jet dy =0. Solve the initial-value problems in Exercises 15-20, In (yt adet etd =~ (et devas 13) 74 1. Gx + By? + Ade Aye? + 5x4 Ody =O, y(t) =2. i teotyteredtetyen (5)=%48 FIRST-ORDER EQUATIONS FOR WHICH XACT SOLUTIONS ARE OBTAINABLE 18, (a? +3y))dx—2xydy=0, 2m 6 19, (Qx—Sy)dx+f4x—yidy= 0, yl. 20. x? + Sy 4 Sy?) dx — (6x? + 43) dy =0, 21. (a) Show thatthe homogeneous equation y0Q)= 6 (Ax + By) de + (Cx + Dy) dy =0 is exact if and only if B= C. (©). Show that the homogeneous equation (Ax? + Bay + Cy?) dx + (Dx? + Exy + Fy) dy = 0 is exact if and only if B = 2D and E = 2C. 22. Solve each of the following by two methods (se Exercise 21(a) @ &+2)éx+x-ydy =O. 0) Gx—y)ax— (4 Ydy=O. 23. Solve each ofthe following by two methods (see Exercise 21(b)} () (2 + 2y%)de + xy y*)dy =0. (0) Gx? + Dey + 9?) dx + 697 + Dap) dy 24. (a) Provethatif M dx + N dy = Oisa homogencous equation, then the change of variables x = uy transforms this equation intoa separable equation in the variables w and x. (b) Use the resut of (a) to solve the equation of Example 2.12 of the text (6) Use the result of (a) to solve the equation of Example 2.13 of th text 25. Suppose the equation M dy +N dy = 0 is homogeneous. Show that the trans- formation x = r¢0s 0, y =r sin 8 reduces this equation to a separable equation in the variables r and 8, 26. (@) Use the method of Exercise 25 to solve Exercise 8 (©) Use the method of Exercise 25 to solve Exercise 9 27. Suppose the equation Mdx+Ndy=0 a is homogeneous. (@) Show that Equation (A) is invariant under the transformation weak’ 8 where kis a constant. (8) Show that the general solution of Equation (A) can be written inthe form aneo(2). © whore cis an arbitrary constant.f-S ee aee a egeeeee CeCe 23 LINEAR EQUATIONS AND BERNOULII Equations 49 (6) Use the result of) to show that the solution (C)is also iny transformation (B), (@)_ Interpret geometrically the results proved in (a) and (0). jant under the 2.3 LINEAR EQUATIONS AND BERNOULLI EQUATIONS, ‘A. Linear Equations In Chapter 1 we gave the definition ofthe linear ordinary differential equation of order 1m; we now consider the linear ordinary differential equation of the fist order. DEFINITION A first-order ordinary differential equation is linear in the dependent variable y and the independent variable x if iti, oF can be, written in the form 4 2 mioy = 063 29 For example, the equation ¥ . Ett Dyas isa Srstorder liner diferential equation, for it can be written as dy ( Lh eer) dea(is pews ‘which is of the form (2.26) with P(x) = 1 + (1/x) and Ql Let us write Equation (2.26 in the form [Posy ~ Qtx)] dx + dy = 0. e2n Equation (2.27) is of the form M(x, y) dx + N(x, 9) dy = 0, where M(x, 3) = PO)y~ OG) and Mx,» Since M(x, 9) a Equation (2.27) is nor exact unless P(x) = 0, in which case Equation (2.26) degenerates into a simple separable equation. However, Equation (2.27) possesses an integrating factor that depends on x only and may easily be found, Let us proceed to find i. Let us multiply Equation (2.27) by p(x), obtaining Ca PCy — Ho O(8I] dx + nls) dy = 0. 228) aN.) Pts) ang SMG?50 FRST.ORDER EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE By definition, u(x) is an integrating factor of Equation (228) if and only if Equation (2.28) is exact; that i, if and only if a a rr Cals) Py = nX)QG)) = & Tue) “This condition edues to a MLpP(x) = a Cat). (229) Jn 229), Pisa boown function ofthe independent variable x, bt isan unknown function of x that we are trying to determine. Thus we write (2.29) as the differential vaoaton HP(x) = x in the dependent variable j: and the independent variable x, where P is a known function of x. This differential equation is separable; separating the variables, we have Pis)de. Integrating, we obtain the particular solition nia) ~ freee aa erne 2.30) where itis clear that 4 > 0. Thus the linear equation (2.26) possesses an integrating factor of the form (2.30). Multiplying (2.26) by (2.30) gives inne dd gare my = gee, which is precisely 4 peitondeyy = Phd leroy] = Dee! Jegrating tha we obtain the solution of Equation (2.26) in the form eirenty = f ATMO de + 6 where cis an arbitrary constant Summarizing this discussion, we have the following theorem THEOREM 2.4 The near diferental equation 2 + PCO = Ole (226)23. UNEAR EQUATIONS AND ataNOULL Equations 51 hasan megan factor of the form A one-parameter family of solutions of this equation is yet fanongey acs tha, sarimal farm ee +] Furthermore itcan be shown that this one-parameter foil of solutions ofthe linear equation (2.26 includes all solutions of 2.26) ‘We consider several examples > Example 2.14 2.74). a3) ty)! ‘and hence an integrating factor is on, f Pe | = on fi ons \ae] expQx + In|x)) = exp(2x) exp(in [xl) = x exp(2s)* ng Equation (231) through by this integrating factor, we obtai xem Ds ears nyex 4 ety) = Roeax Integrating, we obtain the solutions xery Ste yah eS where cis an arbitrary constant. The expresions reap # 80 dene52 esr-onvex EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE > Bxample 2.15 Solve the initial-value problem that consist ofthe differential equation 24% CDR drye x 32) ‘and the initial condition y@)=1. 033) The differential equation (2.32)is not inthe form (2.26) We therefore divide by x? + 1 toobiain ay de x ate a oo Equation (234) is inthe standard form (2.26), where Am integrating factors esol 200d] = exe( f 2) —expance + net = 02+ 1% Multiplying Equation (2.34) through by this integrating factor, we have ct eZ sare s Ny axe? +) or te + yee + Applying the initial condition (2.33), we have S=6 +e. ‘Thus ¢ = 19 and the solution of the initial-value problem under consideration is ox (2 + Dy = +19. > Example 2.16 Consider the differential equation Wax + (xy — Ndy=0. (2.35)123 LINEAR EQUATIONS AND seeNoutet equations 53 Solving for dy/dx, this becomes a iy which is clearly not linear in y. Also, Equation (2.35) is not exact, separable, or ‘homogencous. [t appears fo be of @ type that we have not yet encountered; but let us oak a litle closer. In Section 21, we pointed out that in the differential form ofa first- ‘order differential equation the roles of x and y are interchangeable, in the sense that either variable may be regarded as the dependent variable and the other as the independent variable. Considering differential equation (2.35) with this in mind, let us now regard x as the dependent variable and y as the independent v interpretation, we now write (235) in the derivative form a or a3 t Felee sy 236) ayy ‘Now observe that Equation (2.36) of the form ax Ft Poe = Ov) and 50 is linear in x Thus the theory developed in this section may be applied to Equation (2.36) merely by interchanging the roles played by x and y. Thus an inte- erating factor is os f Py) ]- eo( [2 4) = expla [y?) = 9° Muluipying (2.36 by y* we obtain 24s ys, PB sayteny ton furaes Integrating, we find the solutions in the form vee where cis an arbitrary constant,54 FIRST-ORDER EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE B. Bernoulli Equations ‘We now consider a rather special type of equation that can be reduced 1 a linear ‘equation by an appropriate transformation. This isthe so-called Bernoulli equation. DEFINITION An equation of the form 4 Get Pony = Oy" an is called a Bernoulli differential equation, We observe that ifn = O0r 1, then the Bernoulli equation (2:37) actually a linear equation ands therefore readily solvable as such. However, inthe general casein which #0 ot t this simple situation does aot hold and we raust proceed in a different ‘manner, We now state and prove Theorem 75, which gives a method of solution in the general case. THEOREM 2.5 Suppose n #0 or 1, Then the transformation v = y'~" reduces the Bernoulli equation 24 Pun our ean 10. linear equation in Peoot. We first maltiply Equation (2.37) by y~*, thereby expressing it in the equivalent form + Poay'-* = O60. 23 it welete = yt°% then and Equation (2.8) transforms into ett + Pe = 00) oF, equivalently, E+ (1 — mPlato = (1 — n QC. # a= Pine , Letting PA) = — 9 Pe)23, LINEAR EQUATIONS AND seeNOU EQUATIONS 55, and 2,08) 1 = mow, this may be written de Fe + Pleo Quen ‘which is linear in QED. > Example 2.17 dy 3 Byars 039) ‘This is a Bernoulli differential equation, where m = 3. We first multiply the equation through by y~?, thereby expressing it in the equivalent form 3B yo yieytex 98 then do/dx = ~2y"(dy/d) and the preceding diferent If we let» = y' equation transforms into the linear equ: (240) ‘We see that an integrating factor for this equation is ‘Multiplying 2.40) by e°*, we find Integrating, we find eMpm de MOx + N+, paxtdtce™, Where cis an arbitrary constant. But56. rst-onnen EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE Thus we abtain the solutions of (2.39) in the form 1 pret to Note. Consider the equation Ody - dpi * POM) = Obs, ean where f is known function of y, Letting » = f(y), we have dv _dudy _ afty) dy Bade dy ae and Equation (2.41) becomes & E+ Pode = Ot), ‘which is linear in v. We now observe that the Bernoulli differential equation (2.37) is special case of Equation (241), Weiting (2.37) in the form “y 2 + Poay'-*= 060 ‘and then multiplying through by (1 ~ n), we have = my + Roast = O00, = (1 ~ 1) PC) and Q,(2) = (1 ~ n)Q(2). This is ofthe form (241), where "setting v = py! it becomes é Ft Pulao = O40), which is linear in v. For other special cases of (2.41), see Exercise 37. Exercises ‘Solve the given differential equations in Exercises 1-18, ty yg «dy 1 BBL xt Bs rete B+ = oe 2 xt Ba etyat y age ay 3. Pa syearte 4 Badr = be. ax Pee ieaeeectes 5 Ee 6 W + DF + dw =u dy eeteee ee eeee tn neeeaa 23, UNEAR EQUATIONS AND BEENOULU EQUATIONS 57 8 wax sae ny 9. xdy tbe ty Dax=0, 10. yde + (xy? + x—y)dy=0. ar Spt tan @ + 00s 8. 12 cos r+ (rind ~ cost 40 =0 13, (eos? x — y cos x) dx — (1 + sin x) dy 14, (ysin 2x 00s x) de + (1+ sins) dy =0 yyy ay i, Be 6 xBey dented 1. dy Ay 8y Yd = eee IT dy + (4p — 8y"*)x dx = 0. 18. apt ar Solve the initial-value problems in Exercises 19-30. 19. xB apart, yQ) =8. i Haseyest yee 2. eLy-He + MDa t(er+ Ndy=0, Hh Bayt Ndx—GP 4 Ydy= y= 5. a Sermonow'a (§)=t seers af rt Bayete, gle 2. Bare sea, woae fea- FP 9555 sone a Baye sia, woe sean fh OSES oye Be Baya soa, we sin {Oe 235% yo 2x, O0 every solution approaches 0 as x -» co Consider the differential equ dy i Ft Poly=0. (2) Show that if f and g are two solutions of this equation and c, and e, are eee rset, ec Pa is also a solution of this equation. Consider the diferent! equation 2s royna, “ where P is continuous on a real interval I. (8) Show that the function f such that f(x) = 0 forall x€ i solution of this ‘equation. (&) Show that if fis a solution of (A) such that f(x) = Ofor some xg € J, then F(8) = Ofor all xe I (0 Show thatif f and gare two solutions of (A)such that f(s) = g(Xo)for some Xq€ f then f(x) = g(x) for all x € 1 (a) Prove that iff and g are two different solutions of 2 + Poy = 00. “ then f — gis a soltion of the equation 24 Poyy=o. {0} Tausshow thatiff and gare two different solutions of Equation(A)and cis an arbitrary constant, then ats js one-parameter family of solutions of (A. (@) Let f, bea solution of * Fi+ Poly = 0409)123 LINEAR EQUATIONS ANO seRNoULL! EQuarions 59. and f, be a solution of ay Zt Pedy = Ole), where P, Q,, and Q, are all defined on the same real interval J. Prove that Si + fis a solution of 2+ Peoy=ouy+ O00 onl. (0) Use the result of (a) to solve the equation 84 yezsina tssin2e 36, (@) Extend the result of Exercise 35(a) to cover the case ofthe equation 4 a Ft Poay= F oven, where P, Qx(k = 1,2,....0) are all defined on the same real interval I. (6) Use the result obtained in (a) to solve the equation dy Zey 37. Solve each of the following equations of the form (2.41): dy Jt (0) cosy 2 4 siny = ‘ © OFZ + x07 +29 38, The equation 2% ate? + Bey + C0) w is called Riccat's equation. (@)_ Show that if A(x) =0 for all x, then Equation (A) is a linear equation, ‘whereas if C(x) = 0 forall x, then Equation (A) isa Bernoulli equation, (®) Show that iff is any solution of Equation (A), then the transformation 1 yeasty reduces (A) to a linear equation in. In each of Exercises 39-41, use the result of Exercise 38(b) and the given solution to find a one-parameter family of solutions of the given Riceati equation: ” 2 a 39, Bad 2)? +x — yy — xigiven souton f69)= 160 'RST-ORDER EQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE ay 40, _y? + xy + I; piven solution f(x) 41 Bay? + eld + Dy —(Be? + 4x? — 2): given solution f(x) = x Exercises: Miscellaneous Review Solve each ofthe differential equations in Exercises 1-14, Several can be solved by at least wo different methods. 1. 6xtpde~( + aya. 2 Gxly? — a)dy + Oxy? ~y)de #0. 3B (y= Dade tafe + dy = 4, 7 = 2y)dx—xdy=0. 5. Gx-Sy)dx tet dy =O. 6 ey dx + (ey — 2) dy =O. 1 Oxt'y— 2x) dx + ot + dy =0. 8 Ox bay 4 y)de +28? dy = 0. dy _4xty? —302y Sag a iey dy 0 writs a u. 2 a Itsy? & Bye Solve the inital value problems in Exercises 15-24, 1, G+ yTdx=2xydy =O, HD 16 2? H4de + —x Awd =O, y)=0. 17. (ey = 2x)dxteMydy=0, (=. 18 Ge? +20") dx + Qxy4 6y")dy=0, =? B tw Baten @=| 224 src wreceaTinG FACTORS AND Teansronuarions 61 y= 2. nT 92 2 Dyas, where f= { 1, OSx<2, , >, 30) dx, O Example 2.20 (e+ 2y43)de4 Ox tay Nay =0. 256)414 SPECIAL INTEGRATING FACTORS AND TRANSFORMATIONS 67 Here, = 1,6, = 24 =2,D, = 4and y/o =bs/b, “Theorem 27, We therefore let Therefore, this is Case 2 of pax42y, and Equation (2.56) transforms into t+ nderar0(54 =o Tdx + (22 I)de=0, Which is separable, Integrating, we have Jett Replacing z by x + 2y we obtain the solution of Equation (2.56) in the form Txt e+ DP (e+ IW Re XP day + dy? 4 6x = Dy C. Other Special Types and Methods; An Important Reference “Many other special types of first-order equations exist for which corresponding special ‘methods of solution are known, We shall nat go into such highly specialized types in this book. Instead we refer the reader to the book Differentialgleichungen: Losungsmethoden und Losungen, by E. Kamke (Chelsea, New York, 1948). This re ‘markable volume contains discussions of a large number of special types of equations and their solutions. We strongly suggest that the reader consult this book whenever he ‘encounters an unfamiliar type of equation. Of course one may encounter an equation {or which no exact method of solution is known. In such a case one must resort to ‘various methods of approximation. We shall consider some of these general methods in Chapter & Exercises Solve each differential equation in Exercises 1-4 by fist finding an integrating factor 1 Gay + 4y* + 1) dx + (2? + Dey) dy =O, 2% Qe ttany) de +(x — x? tan y)dy = 0. 3. [e+ I+ y)dx + Oxy + Ody 4 Gy + yd + Oy 9d In each of Exercises $ and 6 find an integrating factor of tue form x) and solve 5. (xy? + Oy) dx + (Sxy + Bx) dy =0, 6 (xy? —2y4) dx + (5x29? — Buy) dy =O.HRST-ORDER KQUATIONS FOR WHICH EXACT SOLUTIONS ARE OBTAINABLE Solve each differentia equation in Exercises 7-10 by making suitable transformation, 71. Gx42y+ Ndx4Qxty + dy =0, 8 Gx-y+ Ide (6x -2y- dy =0. 9. 2 de + Ox +y— dy 10, (10x ~4y + 12)dx— (e+ Sy dy =O. Solve the initial-value problems in Exercises 11-14 LL (6x4 4y + dx + Gx + 2y + dy vi) 12 Gro y-Oded(x+y42dy=0, yQ)=-2. 1B Ort 3yt Ddx txt Ot dy 0, y(-2)—2 1 Gxt apt de diet ys Ddy=O, yO)= 4 IS, Prove Theorem 26. 16, Prove Theorem 2.7. 17. Show that if (x,y) and o(, y) are integrating factors of M(x») dx + Nx, 9) dy w such that u(x, y/o(s, 9 is not constant, then bs, 9) = cobs») is a solution of Equation (A) for every constant ¢ 18. Show that if the equation Mos, 9) dx + Nx yb dy “ay is homogeneous and M(x, y)x + N(x, y)y #0, then 1/[M(x, yx + N(x yh] is an integrating factor of (A). 19, Show that ifthe equation M(x, »)dx + N(x, y) dy = is both homogeneous and exact and if M(x, y)x + Nix, )y is not a constant, them the solution of this, equation is M(x, y)x + NIX, yy = €, where cis an arbitrary constant. 20. An squation that is of the form ya px> flv “ay where p = dy/dx and / is a given function, is called a Clairaut equation. Given ‘such an equation, proceed as follows: 1. Differentiate (A) with respect to x and simplify to obtain t+ son 20 (8) Observe that (B) is fstorder diferental equation in sand p 2, Assume x + /"(p) # 0, divide through by this factor, and solve the resulting ‘equation to obtain, pee © where cis an arbitrary constant\ | | i 21 124 SPECIAL INTEGRATING FACTORS AND TRANSFORMATIONS 69 3. Eliminate p between (A) and (C) to obtain peer 4 fle o ‘Note that (D) is one-parameter family of solutions of (A) and compare the form of dilferential equation (A) withthe form of the family of solutions (D) 4. Remark Assuming x + /"(p) = Oand then eliminating p between (A) and x-+ #0) = O may lead to an “extra” solution that is not a member of the one parameter family of solutions of the form (D). Such an extra solution is usually called a singular soluion, For a specific example, see Exercise 21. Consider the Clairaut equation eas & yopet py whee po. (a) Find a one-parameter family of solutions of this equation. {b) Proceed asin the Remark of Exercise 0 and find an “extra” solution that is, ‘not a member of the one-parameter family found in part (a). ph the integral curves corresponding to several members of the one- parameter family of part (a); graph the integral curve corresponding to the “extra” solution of part (bj, and describe the geometric relationship between the graphs of the members ofthe one-parameter family and the graph of the “extra” solution. ©== CHAPTER THREE—— 3a Applications of First-Order Equations {In Chapter 1 we pointed out that differential equations originate from the mathemati~ cal formulation of a great variety of problems in science and engineering. In this ‘chapter we consider problems that give rise to some ofthe types of first-order ordinary differential equations studied 18 Chapter 2. Furst, we formulate the problem mathemat- icaily,ithereby obtaining a differential equation. Then we solve the equation and attempt to interpret the solution ia terms of the quantities involved in the original problem. ORTHOGONAL AND OBLIQUE TRAJECTORIES A. Orthogonal Trajectories DEFINITION Let Fone) =0 on be a given one-parameter fanily of curces Inthe xy plane. A curee that intersects the ‘eurves ofthe family (3.1) at right angles is ealled an orthogonal trajectory ofthe given family D xample 34 Consider the family of circles way 2)| | | | | i \ | i 21 ORTHOGONAL AND OsLIQue reawcrowes 71 with center at the origin and radius c. Bach straight line through the origin, yok, 63) isan orthogonal trajectory ofthe family of circle (32). Conversely, each circle of the family @2}isan orthogonal trajectory of the family of straight lines (3.3). The families (G2) and (.3)are orthogonal trajectories ofeach other. In Figure 3.) several members of the family of circle (3.2), drawn solidly, and several members of the family of straight lines (3.3), drawn with dashes, are shown, The problem of finding the orthogonal trajectories ofa given family of curves arises inmany physical situations. For example, ina two-dimensional lect ied the ines of force ux lines) and the equipotental curves are orthogonal trajectories ofeach other. ‘We now proceed to find the orthogonal trajectories of a family of curves Fix, =0. en ‘We obtain the differential equation of the family (3.1) by irs differentiating Equation (3.1) irmplicitly with respect to” and then eliminating the parameter c between the derived equation so obtained and the given equation (3.1) itself. We assume that the resulting diferential equation of the family (3.1) ean be expressed in the form 4 Fl 9h G4) ae Thus the curve € of the given family (31) which passes through the point (x ) has the slope f(xs9) there, Since an orthogonal trajectory of the sven family intrsets cach curve of the family at ight angles, the slope ofthe orthogonal trajectory to Cat bss easel Fey)APPCATIONS OF FRST.ORDER EQUATIONS Thus the differential equation of the family of orthogonal trajectories is aw 4 ci ees 5 a” Few te A one-parameter family Gx, ya=0 Te) of solutions of the differential equation (3.5) represents the family of orthogonal trajectories ofthe original family (3.1) except possibly for certain trajectories that are vertical lines. ‘We summarize procedure as follows: Procedure fo Finding the Orthogonal Trajectories of a Given Family of Curves Step 1. From the equation Fou y=) en of the given family of curves, find the differential equation 4 Zo joy 6a) ofthis family, Suep 2. In the differential equation dy/dx = f(x, y) so found in Step 1, replace f(s») by its negative reciprocal ~1/ fx,» This gives the differential equation aye 2-75 69 of the orthogonal trajectories, Step 3. Obtain a one-parameter family G{x,y0)=0 of y= Foxe) of solutions of the differential equation (3.5), thus obtaining the desired family of ‘orthogonal trajectories (except possibly for certain trajectories that are vertical lites and must be determined separately) Caution. In Step 1, in finding the differential equation (3.4) ofthe given family, be sure to eliminate the parameter c during the process. > beample 3.2 Tn Example 3.1 we stated that the set of orthogonal trajectories ofthe family of circles 6a31 ORTHOGONAL AND OBLIQUE TRAKCIORKS 73 isthe family of straight lines yoke. 63 Let us verify this using the procédure outlined above Step 1. Differentiating the equation Sayiad 2) of the given family, we obtain sifleo From this we obtain the differential equation ay ky of the given family (3.2) (Note thatthe parameter cwas automatically eliminated in this case) 66) Step 2. We replace —x/y by its negative reciprocal y/x inthe differential equation (G.6)t0 obtain the differential equation dy vy zr an (of the orthogonal trajectories. Step 3. We now solve the differential equation (3.7). Separating variables, we have dy _ ae y integrating, we obtain yaks, 63) ‘his is @ one-parameter family of solutions of the differential equation (3.7) and thus represents the family of orthogonal trajectories of the given family of circles (32) (except for the single trajectory that is the vertical line x =O and this may be deter- ‘mined by inspection) > Example 3.3 Find the orthogonal trajectories of the family of parabolas y = ex? Step 1. We first find the differential equation of the given family yee. 63) Differentiating, we obtain 69)74 armucarions 0F siest-0RDER EQUATIONS Figure 3.2 Eliminating the parameter c between Equations (3.8) and (3.9), we obtain the differ. ‘ential equation of the family (2.8) in the form ay ik 6.10) Step 2. We now find the differential equation of the orthogonal trajectories by replacing 2y/x in (3.10) by its negative reciprocal, obtaining, dy x ky Ga ‘Step 3. Wenow solve the differential equation (2.1). Separating variables, we have 2y dy = ~x dx. Integrating, we obtain the one-parameter family of solutions of (3.11) in the form. ewer, where kis an arbitrary constant. Thisis the family of orthogonal trajectories of (3.8); it is clearly a family of ellipses with centers at the origin and major axes along the x axis. ‘Some members of the original family of parabolas and some of the orthogonal {tajectories (the ellipses) are shown in Figure 3.2. B. Oblique Trajectories 7 DEFINITION Let Fix,y.)=0 3.12)I a 3.1. ORTHOGONAL AND OBLIQUE TRAECTORES 75 ‘eaone-parameter family of eurves. A curce that intersects the curves of the family(3.12) ‘ata consrant angle 2 # 99° is called an oblique trajectory of the given family. ‘Suppose the differential equation of a family is “y Besos 61) ‘Then the curve of the family (3.13) through the point (x, y)has slope f(x, y) at (x, y)and hence its tangent ine has angle of inclination tan~*L/(x, 9] there. The tangent line of an obique trajectory that intersects this curve at the angle awl thus have angle of incination tan7'Lflx y+ a the point (x) Hence the slope of this oblique trajectory is given by (uy) + tana 7% yan @ ‘Thus the differential equation of such a family of oblique trajectories i given by ay | flowy) rtana de 1 fix pian ‘Thus to obtain a family of oblique trajectories intersecting a given family of curves at the constant angle a # 90°, we may follow the three steps inthe above procedure (page 72) for finding the orthogonal trajectories, except that we replace Step 2 by the fol owing step: tan{in*C sts. +3} Step 2. In the diferental equation dy/dx = fxs) of the given family, replace FG, 9) by the expression Sou y)+ tan 17 han 44) T= fle jana 14) > Example 3.4 Find a family of oblique trajectories that intersect the family of straight lines y = cx at angle 45" Step 1, From y = cx, we find dy/dx = c, Eliminating ¢, we obtain the differential equation oy gaz cus of the given family of straight lines. Step 2, We replace f(x») = 9) Eauation (15) by Se yittana | yixt+l_xty ‘Fix stan76 [APPUICATIONS OF FIRST-ORDER EQUATIONS (tan a= tan 45° = 1 here), Thus the differential equation of the desired oblique trajectories is dy _xty ae x=) 16 Step 3. We now solve the differential equation (3.16). Observing that it is a homogeneous differential equation, we let y = vx to obtain Integrating we obtain 4 lao? + 1) — arctan » = —In |x} ~In Je} In c2x?(o? + 1) ~ 2 arctan v = 0. Replacing v by y/x, we obtain the family of oblique trajectories in the form 7 In c%(x? + y?) —2 arctan = = 0. Exercises In Exercises 1-9 find the orthogonal trajectories of each given family of curves. In each case sketch several members of the family and several of the orthogonal trajectories on the same set of axes. 2 4 6 8 10. Find the orthogonal trajectories of the family of ellipses having center at the ‘origin, a focus atthe point (c, 0), and semimajor axis of length 2c 11, Find the orthogonal trajectories of the family of circles which are tangent tothe y axis atthe origi, 12, Find the value of K such that the parabolas y = ¢,x? + K ate the orthogonal trajectories of the Family of ellipses x? + 2)? — y = cy32 promws in mecHanics 77 13, Find the value of m such that the curves x" + y"= cy are the orthogonal tuajectories of the family 14. A given family of curves is said to be seiforthogonal iit family of orthogonal trajectories is th same asthe piven family. Show that the family of parabolas yi = 2ex + cf isselForthogonal. 15, Find a family of oblique trajectories that intersect the family of circles x*+y* at angle 4 16, Finda family of oblique trajectories that intersect the family of parabolasy? = ex at angle 60° 17, Find a family of oblique trajectories that intersect the family of eurves x+y ‘ex? atfangleasuch that tan a = 2 3.2 PROBLEMS IN MECHANICS A. Introduction Before we apply our knowledge of differential equations to certain problems in mechanics, let us briefly recall certain principles of that subject. The momentum of a ‘body is defined to be the product mv of its mass m and its velocity v. The velocity v and hence the momentum are vector quantities. We now state the following basic aw of mechanics: Newton's Second Law. The time rate of change of momentum of a body is proportional to the resultant force acting on the body and is in the direction of this resultant force. Tn mathematical langage, this law states that KF, a Slo) = KF, ‘where mis the mass of the body, visits velocity, Fis the resultant force acting upon it, and K is a constant of proportionality. If the mass m is considered constant, this reduces to “ eakE, om Pete ous78 serucanons oF naSToROER CQUATIONS where k = 1/K and a = dvjdtis the acccleration of the body. The form 3.17} sa direct statement of the manner in which Newton's second law is usually expressed in words, the mass being considered constant. However, we shall make use of ‘the equivalent form (3.18). The magnitude of the constant of proportionality kdepends ‘upon the units employed for force, mass, and acceleration. Obviously (he simplest systems of units are those for which k = 1, When such a system is used (3.18) reduces to Fema G19) Itisin this fort that we shall use Newton's second law, Observe that Equation (3.19)is 4 Yector equation. Several systems of unitsfor which k= 1 are inuse In this text we shall use only three the British gravitational system (British), the centimeter-gram-second system (eg), and the meter-kilogram-second system mks) We summarize the various units of these three systems in Table 3. Recall that the force of gravitational attraction that the earth exerts on a body is called the weight ofthe body. The weight, being a force, is expressed inforce units. Thus the British system the weight is measured in pounds: in the ces system, in dynes; and the mks system, in newtons. Let us now apply Newtons’s second Taw to a freely falling body (a body falling twward the earth the absence of air resistance), Let the mass ofthe body be m an let w denote its weight. The only force acting on the body i its weight and so this is the resultant force. ‘The acceleration is that due to gravity, denoted by g, which is approximately 32 t/sec? in the British system, 980 em/sec* in the cas system, and 98 m/sec? in the mks system (for points near the earth's surface). Newton's second law F = ma thus reduces to w = mg. Thus ae (2) 2 iF (3.20) 4 relation that we shall requenty employ. Letus now considera body Bin rectilinear mation, that iin motion along straight line On Ewe choose afte reference point as origin O, a fied direction as pos tiveanda unto distance. Then the coordinate xf the postion of Brom the origin O tcllsus the citance or displacement of 8. (Ste Fighre 3.3) The instantaneous velocity of Bis the time rate of change of » ds a TABLE 3.1 Brivish System —__egs System mks System force pound ‘dye newion mass slug gram Kilogram distance foot centimeter meter second second second Rsec? ‘emjsec? m/sec?32 prostims in mechanics 79 Figure 3.3 and the instantaneous acceleration of Bis the time rate of change ofc: @x aa ‘Note that x,t, and a are vector quantities. All forces, displacements, velocities, and ‘accelerations in the positive direction on L ate positive quantities; while those in the negative direction are negative quantities. I we now apply Newtnn's second law F = ma to the motion of B along L, noting, that de do ds. ae aaa ‘we may express the aw in any of the following three forms: do mGnk 620 os ni 0.23) a mf = F, 823) where F isthe resultant force acting an the body. The form to use depends upon the way in which F is expressed. For example, i F isa function of time ¢ only and we desire to ‘obtain the velocity vasa function of 1, we would use (3.21); whereasif Fisexpressed asa fonetion of the displacement x and we wish to find » as a function of x, we would ‘employ (3.23). 1, Falling Body Problems We shall now consider someexamples of a body falling through air toward the earth. In such s ciccumstance the body encounters air resistance as it falls. The amount of air ‘resistance depends upon the velocity of the body, but no general law exactly expressing this dependence is known. In some instances the law R = ko appears to be quite satisfactory, while in others R = ko* appears to be more exact. In any case, the con- * stant of proportionality & in turn depends on several circumstances, In the ex- amples that follow we shall assume certain reasonable resistance laws in each case. ‘Thus we shall actually be dealing with idealized problems in which the true resis- tance law is approximated and in which certain comparatively negligible factors are distegarded.BO srrucarions oF rmst-oRDeR EQUATIONS > Example 35 A body weighing 8 Ib falls from rest toward the earth from a great height. Asit fall, air resistance act upon it, and we shall assume that this resistence (in pounds) is ‘numerically equal to 2c, where vis the velocity (in feet per second), Find the velocity and distance fallen at time 1 seconds, Formulation. We choose the positive x axis vertically downward along the path of the body Band the origin atthe point from which the body fell. The forees acting on the body are: 1. Fits weight, 8 1b, which acts downward and hence is positive. 2. Fy the air resistance, numerically equal to 2c, which acts upward and hence is the negative quantity ~ 20 ‘Soe Figure 34, where these forces are indicated, Newton's second law, F = ma, becomes 3.24 Since the body was initially at rest, we have the initial condition (0) = 0. 025) a | Ren a A 7 anh Figure 3.4\ } | a2 pronews in mecnawes 81 Solution. Equation (3.24 is separable. Separating variables, we have do ort Integrating we find —}1n|8 20] = 4+ e6, which reduces to 8-2 ce Applying the condition (3.25) we find c, = 8. Thus the velocity at time ¢is given by =a, 629 ‘Now to determine the distance fallen at time , we write (3.26) in the form ‘and note that x(0) = 0. Integrating the above equation, we obtain (e+ be") +e, Since x = O when « = 0, we find cy = ~} and henoe the distance fallen is given by atthe ®— 0). G27 Interpretation of Results. Equation (3:26) shows us that as t+ co, the velocity o approaches the limiting velocity 4ft/sec. We also observe that this limiting velocity is approximately attained ina very short time. Equation (3:27) states that as ¢~ 0, x also = 20, Does this imply that the body will plow through the earth and continue forever? ‘Of course not; for when the body reaches the earth’s surface its motion will certainly cease. How then do we reconcile this obvious end to the motion with the statement of Equation (327) Itis simple; when the body reaches the earths surface, the differential equation (324) and hence Equation (3.27) no longer apply? Example 3.6 A skydiver equipped with parachute and other essential equipment falls from rest toward the earth. The total weight of the man plus the equipment is 160 Ib. Before the parachute opens, the air resistance (in pounds) is numerically equal to», where vis the velocity (in feet per second). The parachute opens $ sec after the fall begins; after it ‘opens, the air resistance (in pounds) is numerically equal to fv’, where vis the velocity {in feet por second). Find the velocity of the skydiver (A) before the parachute opens, and (B) after the parachute opens. Formulation. We again choose the positive x axis vertically downward with the ‘origin atthe point where the fall began. The statement of the problem suggests that We break it into two parts: (A) before the parachute opens; (B) after it opens. We first consider problem (A). Before the parachute opens, the frees acting upon the skydiver are: 1, F;, the weight, 160 Ib, which acts downward and hence is positive.82 arrucanions oF nest-oxpen FQUATIONS 2. Fa the air resistance, numerically equal tof which acs upward and hence i the nogative quantity —d. We use Newton's second law F v= ma, where F = F, + Fy let m= w/a, and take 2. We obtain Since the skydiver was initially at rest, = 0 when ¢ = 0, Thus, problem (A), concerned ‘with the time before the parachute opens, is formulated as follows: 028) G29) ‘We now turn to the formulation of problem (B). Reasoning as before, we see that after the parachute opens, the forces acting upon the skydiver are: 1. F, = 160, exactly as before. 2. Fy = —de? (instead of ~40), ‘Thus, proceeding as above, we obtain the differential equation a Syn WO ie% Since the parachute opens 5 sec after the fll begins, we have v = v, when t = 5, where +, is the velocity attained when the parachute opened. Thus, problem (B), concerned with the time afer the parachute opens, is formulated as follows: ao 5G = 100-40, 630) (5) = 0 @3n Solution. We shall frst consider problem (A). We find a one-parameter family of solution of do 55 = 160 ~ fo. 628) Separating variables, we obtain Integration yields Info ~ 320) = —Het +o, which readily simplifies to the form 320 + ce"?——_—— 32 proniens In mecnanics. 83 Applying the initial condition (3.28) that » = Oat ¢ = 0, we find that c = ~320, Hence the solution to problem (A) is v= 32001 — 2°), G32) which is valid for 0 < ¢ < 5. In particular, where ¢ 2) = 126, 833) ‘which is the velocity when the parachute opens. ‘Now let us consider problem (B) We ist find a one-parameter family of solutions of the differential equation dy 522 160 — ge 630) ‘Simplifying and separating variables, we obtain de ae Integration yields 1 e=16 Borie ate neato ay nets ee This readily simplifies tothe form on 16 ee om) and solving this for # we obtain aa 35) — ee Applying the initial condition (3.31) that v = v, at ¢ = 5, where n is given by (3.33) and is approximately 126, to (3.34), we obtain He Substituting this into (3.35) we obtain w= lier 0) 639 Interpretation of Results. Let us fist consider the solution of problem (A), given bby Equation (3.32), According to ths, as ¢ +00, v approaches the limiting velocity 320 ffs, Thus if the parachute never opened, the velocity would have been approx imately 320 see atthe time when the unfortunate skydiver would have stuck the ‘earth! But, according othe statement of the problem, the parachute does open 5 sec which is void for 2 5.84 APPLICATIONS OF FASEORDER EQUATIONS after the fall begins (we tacitly and thoughtfully assume $< TT, where TT is the time when the earth is reached!). Then, referring to the solution of problem (B), Equation. (6.36), weseethat ast + 20, v approaches the limiting velocity 16 ft/sec. Thus, assuming, that the parachute opens at a considerable distance above the earth, the velocity is approximately 16 ft/sec when the earth is fnally reached, We thus obtain the-well- ‘known fact that the velocity of impact with the open parachute is @ small raction of the impact velocity that would have occured if the parachute had aot opened. The calculations in this problem arc somewhat complicated, buc the moral is clear: Make certain thatthe parachute opens! . Frictional Forces Ha body moves on a rough surface, it will encounter not only air resistance but also another resistance force due to the roughness of the surface. This additional force is called friccion. Its shown in physics thatthe friction is given by pN, where 1, risa constant of proportionality called the coeficient of friction, which depends ‘pon the roughness of the given surface; and 2. Nisthe normal that is, perpendicular force which the surface exerts on the body. We now apply Newton's second law to a problem in which friction is involved, > eample 37 ‘An object weighing 48 [bis released from rest at the top of a plane metal slide that is inclined 30° to the horizontal. Air resistance (in pounds) is numerically equal to one half the velocity (in feet per second), and the coefficient of friction is one-quarter. |A. What isthe velocity of the object 2 sec after itis released? BI the side i 24 ft long, what is the velocity when the object reaches the bottom? Formulation. The line of motion is along the slide. We choose the origin atthe top and the positive xdirection down theslide. If we temporarily neglect the rition and air resistance, the forces acting upon the object A ate: 1. Ts weight 481, which ats vertically downward: and 2. The normal force, N, exerted by the slide which gets in an upward direction perpendicular tothe slide, See Figure 35} ‘The components of the weight parallel and perpendicular tothe slide have magnitude 48 sin 30° = 24 and 48 c08 3 wi respectively, The components perpendicular fo the slide are in equilibrium and hence the normal force N has magnitude 24/3.a2 mosis i mecranes 85 o Figure 3.5 Now, taking into consideration the friction and air resistance, we see that the forces acting on the abject as it moves along the slide are the following: 1. Fythe component on the weight parallel othe plane, having numerical valu 24. Since this fore act in the positive (downward) direction along the slide, we have A 2. Fthefrctiona fore, having numerical value y= 4(25,/3) negative (upward) direction along the side, we have B= 65. 3. Fy the ar resistance, having numerical value $e: Since» > Oand this also aes in the negative direction, we have 2. 12 thisactsin the We apply Newton's second law P= ma, Here P= +A, + y= 4-6/3 Je and m = w/g = $$ = 4. Thus we have the differential equation =~ 6/3— $0. (337) Since the object is released from rest, the intial condition is (0) = 0. 0.38) Solution. Equation (3.37) is separable; separating variables we have a dr ae 27s Integrating and simplifying, we find vn 48— 25 ~ ce “The condition (3.38) gives e, = 48 — 12/3. Thus we obtain (48 — 12 /3\Q1 =e"), 3.39) Question Ais thus answered by letting 1 = 2in Equation (239), We nd (2) = 48 —12,/3M1 — €°*9) x 1020s)BG armications oF rinst-oRDeR EQUATIONS In order to answer question B, we integrate (3.39) o obtain x48 — 2/3[t 4 30°) + 6, (48 ~12,/3)3). Thus the distance covered at time + is given by = (48 — 12 /FMe + 3 — 3) Since the slide is 24 ft long, the object reaches the bottom atthe time T determined from, the transcendental equation 26 = (48 = 12VINT + 30-7 — 3), which may be writen as Since x(0) = 0,¢5 ate The value of T that satisfies this equation is approximately 2.6. Thus from Equation {.39) the velocity of the object when it reaches the bottom is given approximately by (48 ~ 12/31 — e°) = 123 (sea), Exercises 1. A.stone weighing 4 Ib falls from rest toward the earth from a great height. As it falls itis acted upon by air resistance that is numercally equal to 4v (in pounds}. where vis the velocity (in feet per second), (a) Find the velocity and distance fallen at time ¢ see. (b) Find the velocity and distance fallen at the end of 5 sec. 2 Atal wishing bisthrown vera dowaward toward hearth om aiht Sino wih anata scocly ofc Ass sated pom 9 resistance that is numerically equal to 3n(in pounds), where-vis the velocity (in feet persed (0) Whats the slot and iste fen atthe nd of oe minute? (8) With what vlc des he al sie he art? 3. A ball weighing } Ib is thrown vertically upward from a point 6 ft above the Site of thar than nal voy of 2 /ne utes ced spon by air resistance that is numerically equal to ayo (in pounds), where » is the ‘clon inet per pond), How hgh wl he al sie | 4. A ship which eigh 32000 tons tart om etude he feof constant | propel hut of 00,0 Ib The cance in pooner ea 0 Behr osm et pe sco (0) Find the vey ofthe ship as foto ofthe time (0) Find theming velo (hati thei of vas = +2) (6) Find bow tong aes the ship to atin a velo of 0% ofthe ing | wei | 5. A body of mass 10 gi dropped fom et toward the earth rom a hight of | Too0 mA al ai ven cs upon and ths esas Reto)32 promens in mecHanics 87 proportional to the velocity «(in meters per second). Suppose the limiting velocity is 245 m/sec. (a) Find the velocity and distance fallen at time r secs. (®) Find the time at which the velocity is one-fith of the limiting velocity. 6, Anobject of mass 100 gis thrown vertically upward from 2 point 60 cm above the ‘earth's surface with an intial velocity of 150.em/sec. It rises briefly and then falls, vertically to the earth, all of which time it is acted oa by air resistance that is ‘numerically equal to 2000 (in dynes), where vis the velocity (in em/sec) (a) Find the velocity 0.1 see after the object is thrown, (0) Find the velocity 0.1 sec after the object stops rising and starts falling. 7. Two people ate riding in a motorboat and the combined weight of individuals, ‘motor, beat, and equipment is 640 Ib. The motor exerts a constant force of 20 1b fon the boat in the direction of motion, while the resistance (in pounds) is ‘numerically equal to one and one-half times the velocity (in feet per second). Ifthe boat started from rest, ind the velocity of the boat after (s) 20 sec, b) { min. I ( \ | seeuenaen ena { (a) Find the velocity of the boat 15 sec after the tow rope was cast off. [ \ } i \ I | 9. A bullet weighing 102 is fired vertically downward from a stationary helicopter witha muzzle velocity of 1200 ft/sec. The ar resistance (in pounds) is numerically equal to 167 x, where vis the velocity (in feet per secon), Find the velocity of the billet asa function of the time, 10. A shell weighing 1 Ib is fred vertically upward from the earth’s surface with a ‘muzzle velocity of 1000 ft/sec. The air resistance (in pounds) is numerically equal to 10° *c%, where vis the velocity (in fet per second). (0) Find the velocity ofthe rising shell asa function of the time, (0) How long will the shell rise? 11, An object weighing 16 Ibis dropped from rest on the surface of a calm lake and thereafter starts to sink. While its weight tends to foree it downward, the buoyancy of the object tends to forceit back upward. If this buoyancy force is one of 6 Ib and the resistance ofthe water (in pounds) is numerically equal to twice the square of the velocity (in fei per second), find the formula forthe velocity of the sinking object asa function of the time. 12. An object weighing 12b is placed beneath the surface of a calm lake. The buoyancy of the object is 30 Ib; because of this the object begins to rise. If the resistance of the water (in pounds) is numerically equal to the square of the velocity in fet per second) and the object surfaces in 5 sec, find the velocity of the object at the instant when it reaches the surface.8B szpucaTions oF nstoRDER tQUATIONS 3. 16. ‘A man is pushing a loaded sled across a level field of ice atthe constant speed of 10 ft/sec. When the man is halfway across the ice field, he stops pushing and lets the loaded sied continue on. The combined weight ofthe sled and its load is 80 Ib; the ar resistance (in pounds) is numerically equal to 4, where vis the velocity of the sled (in fet per second); and the coefficient of friction of the runners on the ice is 004. How far will the sled continue to move after the man stops pushing? A girl ondee sled has just slid down a hill onto a level eld of ice and is starting to slow down, AC the instant when their speed is 5 ft/sec, the gir!’s father runs up and begins to push the sled forward, exerting a constant force of 15 Ibn the direction cof motion. The combined weight of the giel and the sled is 96 Ib, the air resistance (in pounds) is numerically equal to one-half the velocity (in feet pet second), and the coefficient of frietion of the runners on the ice is 0.05. How fast isthe sled ‘moving 10 sec after the father begins pushing? ‘A case of canned milk weighing 24 Ibis released from rest at the top of a plane ‘metal slide which is 30 ft fong and inclined 45" to the horizontal. Air resistance (in pounds) is numerically equal to one-third the velocity (in fet per second) and the Coeficent of friction is 0.4 (a) What is the velocity of the moving case 1 sec after its released? (b) What isthe velocity when the case reaches the bottom of the slide? ‘A boy goes sledding down a long 30° slope. The combined weight of the boy and his sled is 721b and the air resistance (in pounds}is numerically equal to twice their velocity (in fet per second). If they started from rest and their velocity atthe end of 5 sec is 10 fi/sec, what is the coefficient of friction of the sled runners on the ‘An object weighing 32 Ibis released fom rest 50 ft above the surface of a calm lake, Before the object reaches the surface ofthe lake, the air resistance (in pounds) is given by 2r, where v isthe velocity (in feet per second). After the object passes beneath the surface, the water resistance (in pounds) is given by 6v. Further, the object is then buoyed up by a buoyancy force of 8b. Find the velocity of the object 2 sec aftr it passes beneath the surface of the lake, AA rocket of mass m is fred vertically upward from the surface of the earth with initial velocity » = v. The only force on the rocket that we consider is the gravitational attraction of the earth. Then, according to Newton's law of ‘gravitation, the acceleration a of the rocket is given by a = —K/x?, where k > 0 §s a constant of proportionality and x is the distance “upward” from the center of the earth along the line of motion. At time ¢ = 0, x = R (where R isthe radius of the earth), a = —g (where g is the acceleration due to gravity), and Express a= dod asin Equation (323), apply the appropriate initial data, and note that v satisfies the differential equation Solve this diferent equation, apply the appropriate inital condition, and thas express vasa function of x In particular, show that the minimum valve of t for which the racket wil escape from the earth ix /29R. This isthe soled tele33 wre rromens 89 ity of escape; and using R= 4000 miles, g = 32 t/sec?, one finds that this is approximately 25,000 mph (or 7 mi/sec. 19. A body of mass mis in rectilinear motion along a horizontal axis, The resultant force acting on the body is given by —kx, where k>0 is a constant of propoctionaity and xis the distance along the axis rom axed point O. The body has initial velocity 0 = v9 when x = xo. Apply Newton's second law in the form (6.23) and thus write the differential equation of motion in the form ao me = hs Solve the differential equation, apply the initial condition, and thus express the ‘square of the velocity «asa function of the distance x, Recalling that v = dx/dt, ‘show that the relation’ between v and x thus obtained is satisfied for alltime ¢ by where ¢ is a constant. 3.3 RATE PROBLEMS In certain problems the rate at which a quantity changes is a known function of the amount present and/or the time, andit is desired to find the quantity itself. Fx denotes the amount of the quantity present at times, then dx/dt denotes the rate at which the quantity changes and we are at once led to differential equation I this section we consider certain problems ofthis ype ‘A, Rate of Growth and Decay > Example 3.8 ‘The rateat which radioactive nuclei decay is proportional (othe number of such nuclei that are present in a given sample, Half of the original number of radioactive nuclei ‘have undergone disintegration in a period of 1500 years 1. What percentage of the original radidactive nuclei will remain after 4500 years? 2. Imhow many years will only one-tenth of the original number remain? Mathematical Formulation. Let x be the amount of radioactive nuclei present “after t years. Then ddr represents the rate at which the nuclei decay. Since the nuclei decay at a rate proportional fo the amount present, we have Gnks G40) ‘where K isa constant of proportionality. The amount xis clearly positive; further, since xisdecreasing, dx/dt < 0. Thus, from Equation (3.40), we must have K <0. Inorderto90 arpucaTions oF riesroanes EQuarions ‘emphasize that x is decreasing, we prefer to replace K by a positive constant preceded —K > Oand write the differential equation (3.40) in ax a Gan Letting xo denote the amount initially present, we also have the initial condition (0) = Xo, 647) We know that we shall need such a condition in order to determine the arbitrary constant that wll apgear in a one-parameter family of solutions of the differential equation (341), However, we shall apparently need something else for Equation (3.41) contains an unknown constant of proportionality k. This “something else" appears in the statement of the problem, for we are told that half of the original number disintegrate in 1500 years. Thus half aso remain at that time, and this at once gives the condition : (1500) = 4x9 643) Solution, The differential equation (3.41) is clearly separable; separating variables, integrating, and simplifying, we have at once Applying the initial condition (3.42), x = xy when t= O, we find that c= x, and hence ‘we obtain 44) We have not yet determined k. Thus we now apply condition (3.43), x = 4xp when = 1500, to Equation (3.44), We find dx = xe, or (erty wt, or finally ete qyinsee, (45) From this equation we could determine k explicily and substitute the result into Equation (344). However, we see from Equation (3.4) that we actually do not need k itself but rather only e~*, which we have just obtained in Equation (345). Thus we substitute e~# from (3.45) into (344) to obtain x= xofe“*Y = xoL(d)8]' xo xo(AV5°0, 46) Equation (3.46) gives the number x of radioactive nuclei that are present at time t ‘Question Tasks us what percentage of the original number will remain after 4500 years. ‘We thus let: = 4500 in Equation (3.46) and find old? = #Xo.aa rare mrontens 91 “Thus, one-eighth or 125% ofthe orginal number remain after $500 years. Question 2 asks us when only one-tenth wll remain. Thus we let x = 4px in Equation 3.46) and solve fort We have wayne Using logarithms, we then obtain In) = Igy” = 5 tn). From this it follows at once that at _ ini 1500 ~ in 1500 in 10 150010 10 3585 years B, Population Growth ‘We next consider the growth of a population (for example, human, an animal species, ‘ora bacteria colony) as a function of time. Note that a population actually increases discontinuously by whole number amounts. However, ifthe population is very large, such individual increases init are essentially negligible compared to the entire pop” ulation itself In other words, the population increase is approximately continuous. We shall therefore assume that this increase is indeed continuous and in fact that the population isa continuous and differentiable fucntion of time. Given a population, we let x be the number of individuals in it at time ¢.1f we assume thatthe rate of change of the population is proportional to the number of individuals init at any time, we are led to the diferential equation ds a where kis @ constant of proportionality. The population xis positive and is increasing and hence dxjdx > 0. Therefore, froi(3:47), we must have k > 0. Now suppose that at time t the population is x9. Then, in addition to the differential equation (3.47), we hhave the initial condition % ean Alto) = Xo. 3.48) ‘The differential equation (3.47) is separable. Separating variables, integrating, and simplifying, we obtain xa cet Applying the intial condition (3.48), x = xpat = to, to this, wehave x9 = ce, From this we al once find e = xye"*® and hence obtain the unique solution x= xqett (3.49) of the differentia equation (3.47), which satisfies the initia condition (3.48)92 arpucations oF ninst-onDER EQUATIONS From (3.49) wesee tha a population governed bythe dilferential equation (347) with k > Oandinitial condition (348) is one that increases exponentially with ime. This aw ‘of population growth is called the Malthusian law, We should now inquire whether or not there are cases in which such a model for population growth is indeed realistic. In ‘answer to this, tan be shown tha this ode), with suitable value of kis remarkably accurate in the case of the human population of the earth during the last several decades (see Problem 8(b)}. It is aso known to be outstandingly accurate for certain mammalian species, with suitable k, under certain realizable conditions and for certain time periods. On the other hand, turning back to the case of the human population of the earth, i can be shown thatthe Malohusian law turns out to be quite unteasonable when applied tothe distant future (se Problem 8()).Itisalso completely unrealistic for other populations (or example, bacteria colonies) when applied over suicientl long periods of time. The reason for ths is not hard to see. For, according to (3.49), a Population modeled by this law always increases and indeed does so at an ever increasing rate; whereas observation shows that a given population simply does not row indefinitely. Population growth s represented more realistically in many cases by assuming that the number of individuals x in the population at time tis described by a diferential ‘equation ofthe form ax : Fm bax, 50) where k > O and 4 > 0 are constants. The additional term ~ x? isthe result of some ‘cause that tends to imit the ultimate growth of the population. For example, such a ‘cause could be insuficient living space or food supply, when the population becomes suificientl large. Concerning the choice of — ix? forthe term representing the effect of the cause, one can argue as follows: Assuming thecause affects theentire population of -xmembers then the effect on any one individuals proportional x. Thus the effect on all x individuals in the population would be proportional to x x = 2°. ‘We thus assume that a population is described by a differential equation of the form {€.50), with constants k > 0 and > 0, and an initial condition of the form (348) In ‘most such cases it turns out thatthe constant 2is very small compared to the constant Thus for sufficiently small x the term kx predominates, and so the population grows very rapidly fora time. However, when x becomes sufficientlylarge, the term —Ax*is of ‘comparatively greater influence, nd the result ofthis is a decrease inthe rapid growth rate. We note thatthe differential equation (3.50) is both a separable equation and a Bernoulli equation. The law of population growth so described is called the logistic law of growth, We now considera specific example of this type of growth. > Example3.9 ‘The population x of a certain.city satisfies the logistic law ee a7 300" ~ Gor as ‘where time ¢is measured in years. Given that the population of this city is 100,000 in 1980, determine the population as a function of time fort > 1980. In particular, answer33 mate rtooums 93. the following questions: (a) What will be the population in 20007 (&) In what year does the 1980 population double? (©) Assuming the differential equation (3.51) applies forall > 1980, how large will the population ultimately be? Solution. We must solve the separable differential equation (3.51) subject to the ini solution ° - x(1980) = 100,000, 3.52) ‘Separating variables in (3.51), we obtain as oreo and hence. dx Worst GH Using partial fractions, this becomes 1 oy" § wales) Integrating, assuming 0 < x < 10°, we obtain: Loon x ~ Ia = (0) I} w+ and hence of apag)eoiete. Thus we find T= 9% ~ Solving this for x, we finally obtain cet00 =a 0s) Now applying the initial condition (3.52) to this, we have 00 = eager i coy? aos | FFT TOT HPF Substituting this value for ¢ back into (3.53) and simplifying, we obtain the solution in94 aerucanions oF restonDes EQUATIONS the form G4) This gies he poptation x as function of ime fort > 1980. ‘We now consider the questions (a), (b), and (c) of the problem, Question (a} asks for the population in he year 2010 Ths we let 2000 in (54 aad obtain it 5085 Question (b) asks for the year in which the population doubles. Thus we let 2 = 200,100 = 2(10)? in G.54) and solve for ¢. We have 200) = yeaa from whic ‘and hence 1 2061 Question 6) asks how large the population will ultimately be, assuming the iferenial ‘equation (3.51) applies forall > 1980, To answer this, weevaluate im x ast» 20 using the solution (3.54 of (3.51). We find 0s Him x = fim atresia = (10% = 1,000,000. C. Mixture Problems We now consider rate problems involving mixtures. A substance $i allowed to flow into acertain mixture ina container ata certain rate, and the mixture is kept uniform by stirring. Further, in one such situation, this uniform mixture simultaneously flows out of the container at another (generally different) rate; in another situation this may not be the case. In either case we seek to determine the quantity of the substance S present in the mixture at time t Letting x denote the amount of $ present at time 1, the derivative dx/dt denotes the rate of change of x with respect 10. IF IN denotes the rate at which Senters the mixture and OUT the rate at which it leaves, we have at once the basic equation ae a from which to determine the amount x of $ at time «, We now consider examples. IN- OUT. 855) D> Example 3.10 ‘A tank initially contains $0 gal of pure water. Starting at time t = 0 a brine containing 2b dissolved salt per gallon flows into the tank atthe rate of 3 gal/min, The mixture33 ware raosiems 95. is kept uniform by stirring and the wellstired mixture simultaneously flows out of the tank at the same rate, L. How much salt isin the tank at any time ¢ > 0? 2 How much salts present at the end of 25 min? 3. How much salt is present after along time? Mathematical Formulation. Let x denote the amount of sat in the tank at time ‘We apply the basic equation (3.55), dx a IN — OUT. ‘The brine flows in atthe rate of 3 gal/min, and each gallon contains 2 Ib of salt. Thus IN = (2 Ibjeal)3 gal/min) = 6 tb/min Since the rate of outflow equals the rate of inflow, the tank contains $0 gal of the ‘mixturest any time t. This 50 gal contains x Ib of sal t time t,and so the concentration cof salt at time tis yx Ib/gal. Thus, since the mixture lows out atthe rate of3 gal/min, we have ouT= & Ia!) ‘eal/min a 5 b/min ‘Thus the diferential equation for x as a function of ris de ok G-5-% 356) Since initially there was no salt in the tank, we also have the initial condition (0) = 0. 657) Solution. Equation (3.56) is both linear and separable. Separating variables, we have dx 3 Wo et Integrating and simplifying, we obtain = 100 + e079, Applying the condition (357), x = Oat « = 0, we find that c= ~ 100. Thus we have x = 100(1 — e°99) 638) ‘This is the answer to question 1. As for question 2, at the end of 25 min, ¢ = 25, and Equation (3.58) gives (25) = 100(1 = e° 4) = 760). ‘Question 3 essentially asks us how much salts present as ¢-+ co. To answer this we let 11+ op in Equation (3.58) and observe that x + 100.96 ArmucaTions OF FinsT-onDeR EQUATIONS, > Example 3.11 ‘A large tank initially contains $0 gal of brine in which there is dissolved 10 1b of salt. Brine containing 2 Ib of dissolved salt per gallon flows into the tank at the rate of 5 gal/min. The mixture is kept uniform by stirring, and the stirred mixture simulta neously lows out atthe slower rate of 3 gal/min. How much salt isin the tank at any time 1 > 0? ‘Mathematical Formulation. Let x =the amount of salt t times, Again we shall use Equation (3.55): & = IN - OUT. Proceeding as in Example 3.10, / IN = (2 Ib/gal(5 gal/min) = 10 {b/min; Aso, once again OUT =(C Ib/gal)(3 gal/min), ‘where C Ib/gal denotes the concentration. But here since the rate of outflow is different from that of inflow, the concentration is not quite so simple. At time ¢ = 0, the tank contains 50 gal of brine. Since brine flows in at the rate of $ gal/min but lows out atthe slower rate of 3 gal/min, there isa net gain of 2 gal/min of brine in the tank ‘Thus at the end of 1 minutes the amount of brine in the tank is 50 +2 gal Hence the concentration at time ¢ minutes is 50+ 2 ‘b/eal, anise ovrag’’ yin aa sie dient cnatonBsomes #20 gy re) Sie here sini 101 of an he nk, we oe ini eondon 0-10 ow Solution. The fferential equation (3.59) is not separable but it is linear. Putting it in standard form,a3 sare reonens 97 we find the integrating factor 3 2 ool feta) + 50)", Multiplying through by this, we have Qt sop $32 + SOx = LOgee + 509 4 t+ 50)%%x] = 10 + 50)" Thus (21 + 50% = 200¢ + 50)°7 +6 or xe A+) 4 Soe Applying condition (3.60), x = 10at t = 0, we find Wo 1004 + ope or 2,00 /3. ‘Thus the amount of salt at any time ¢ > Ois given by e 22,5004/2 x 44 100 — Exercises 1. Assume that the cate at which radioactive nuclei decay is proportional to the ‘numberof such nuclei that are present ina given sample. Ina certain sample 10% of the original number of radioactive nuclei have undergone disintegration in 8 period of 100 years. (a) What percentage of the original radioactive nuclei will remain after 1900 years? (b)_ In how many years will only one-fourth of the original number remain? 2. Acertain chemical is converted into another chemical by chemical reaction. The rate at which the first chemical is converted is proportional to the amount of this ‘chemical present at any instant. Ten percent of the original amount of the first ‘chemical has been converted in 5 min, {@) What percent of the first chemical will have been converted in 20 min? (©) Inhow many minutes will 607% of the frst chemical have been convested?9B armuications oF finsT-oRDER EQUATIONS 3 ‘A chemical reaction converts a certain chemical into another chemical, and the rate at which the first chemical is converted is proportional to the amount of this ‘chemical present at any time. At the end of one kour, 50 gm of the first chemical remain; while atthe end of three hours, only 25 gm remain. @ O) © ‘How many grams of the first chemical were present initially? How many grams of the first chemical will remain atthe end of five hours? In how many hours will only 2 gm ofthe frst chemical remain? ‘A chemical reaction converts & certain chemical into another chemical, and the rate at which the first chemical is converted is proportional to the amount of this chemical present at any time. At the end of one hour, two-thirds kg of the frst chemical remains, while at the end of four hours, only one-third kg remains. @ @) ‘What fraction of the first chemical remains at the end of seven hours? ‘When will only one-tenth of the first chemical remain? Assume that the population of a certain city increases at arate proportional 10 the ‘number of inhabitants at any time. I the population doubles in 40 years, in how ‘many years will it triple? ‘The popula of the city of Bingville increases at a rate proportional to the ‘number of is inhabitants present at any time. If the population of Bingvlle was 30000 in 1970 and 35000 in 1980, what will be the population of Bingvile in 19907 In a certain bacteria culture the rate of increase in the number of bacteri proportional to the number present «® 0 If the number triples in 5 hr, how many will be present in 10 hy? ‘When will the number present be 10 times the number initially present? ‘Assume that the rate of change of the human population of the earth is pro- portional to the number of people on earth at any time, and suppose that this population s increasing atthe rate of 2% per year. The 1979 World Almanac gives the 1978 world population estimate as 4,219 million; assume this figure isin fact correct, @ © @ @ © Using this data, express the human population of the earth asa function of time. ‘According tothe formula of part (a), what was the population of the earth in 19507 The 1979 World Almanae gives the 1950 world population estimate as, 2,510 million. Assuming this estimate is very nearly correct, comment on the ‘accuracy of the formula of part (a) in checking such past populations. According othe formula of part (a, what willbe the population of the earth jn 2000? Does this seem reasonable? ‘According to the formula of par (a), what was the population ofthe earth in 19002 The 1979 World Almanac gives the 1900 world population estimate as, 1,600 million. Ascuming this estimate is very nearly correct, comment on the accuracy of the formula of part (a) n checking such past populations. According tothe formula of part (a), what will be the population of the earth in 2100? Does this seem reasonable?10. 2 14 aa ertmonns 99 ‘The human population of a certain island satisfies the logistic law (3.50) with k= 003, A= 3(10)-*, and time t measured in years. (2) If the population in 1980 is 200,000, find a formula for the population in future years {b) According to the formula of part a), what will be the population in the year 20007 (6) What isthe limiting value of the population as t+ co? ‘This isa genetal problem about the logistic law of growth. A population satisfies the logistic law (3.50) and has xo members at time fo. (a) Solve the differentia equation (3.90) and thus express the population x asa function of. (b) Show that as -+ co, the population x approaches the limiting value kj. (6) Show that dx/dt is increasing if x < k/22 and decreasing ifx > k/2A. (@) Graph x asa function of t for > ty {e) Interpret the results of parts (b), (c), and (d). ‘The human population of a certain small island would satisly the logistic law (3.50), with & = fg, 2 = (10)"%, and ¢ measured in years, provided the annual emigration from the island is neglected. However, the fact is that every year 100, people become disenchanted with island life and move from the island to the mainland, Modify the logistic differential equation (350) with the given kand 2s0 4 fo include the stated annual emigration. Assuming that the population in 1980, is 20,000, solve the resulting initial-value problem and thus find the pop- lation of the istand as a function of time. Under natural circumstances the population of mice on a certain island would increase at a rate proportional to the number of mice present at any time, provided the island had no cats. There were no cats on the island from the beginning of. 1970 to the beginning of 1980, and during this time the mouse population doubled, reaching an all-time high of 100,000 at the beginning of 1980. At this time the people of the island, alarmed by the increasing number of mice, imported a number of cats to kill the mice. Ifthe indicated natural cate of increase of mice was thereafter offset by the work of the cats, who killed 1000 mice a month, how many mice remained at the beginning of 1981? ‘An amount of invested money ssid to draw interest compounded continuously it the amount of money increases ata rate proportional to the amount present Suppose $1000 is invested and draws interest compounded continuously, where the annual interest rates 6 (@) How much money’ will be present 10 years afer the orginal amount was invested? (©) How long will it take the original amount of money to double? ‘Suppose a certain amount of money is invested and draws interest compounded continuously. (a) Ifthe original amount doubles in (wo years, then what isthe annual interest rate?100 APPLICATIONS OF FIRST-ORDER EQUATIONS 1s. Yn. 18 (6) Ifthe original amount increases 50% in six months, then how long willit take the original amount to double? ‘A tank intially contains 100 gal of brine in which there is dissolved 20 ib of salt. Starting at time ¢ = 0, brine containing 3 Ib of dissolved salt per gallon flows into the tank at the tate of 4 gal/min. The mixture is kept uniform by stirring and the ‘well-stired mixture simultaneously flows out of the tank at the satne rate, (a) How much satis in the tank at the end of 10 min? (6) When is there 160 tb of salt in the tank? AA large tank initially contains 100 gal of brine in which 10 Ib of satis dissolved. Starting at ¢ = 0, pure water flows into the tank at the rate of 5 gal/min, The ‘mixture is kept uniform by stirring and the wellstirred mixture simultaneously flows out at the slower rate of 2 gal/min, (a) How much salt is in the tank at the end of 15min and what is the concentration at that time? (b} IF the capacity of the tank is 250 gal, what isthe concentration at the instant the tank overflows? |A tank initially contains 100 gal of pure water. Starting at r= 0, a brine containing 4 Ib of salt per gallon flows into the tank at the rate of 5 gal/min. The ‘mixture is kept uniform by stirring and the well-stirred mixture flows out at the slower rate of 3 gal/min, (a) How much satis in the tank at the end of 20 min? o Seeartamne eaere: ‘A large tank initially contains 200 gal of brine in which 15 tb of salt is dissolved. Starting att = 0, brine containing 4 Ib of salt per gallon flows into the tank at the rate of 35 gal/min. The mixture is kept uniform by stirring and the well-stirred mixture leaves the tank atthe rate of 4 gal/min (a) How much salt is in thé tank at the end of one hour? (b)_How much salt isin the tank when the tank contains only $0 gal of brine? ‘A 500 liter tank initially contains 300 liters of fuid in which there is dissolved 50 gm of a certain chemical. Fluid containing 30 gm per liter of the dissolved chemical flows into the tank at the rate of fiters/min. The mixture is kept uniform by stirring, and the stirred mixture simultaneously lows out atthe rate of 25 lters/min, How much ofthe chemical isin the tank at the instant it overflows? ‘A 200 liter tank is initially full of uid in which there is dissolved 40 gm of a certain chemical. Fluid containing 50 gm per liter ofthis chemical flows into the tank atthe rate of 5 liters/min, The mixture is kept uniform by stirring, and the stirred mixture simultaneously lows out atthe rat of 7 liters/min, How much of ‘the chemica? isin the tank whem itis only half ull? ‘The air im a room whose volume is 10,000 eu ft tests 0.15% carbon dioxide. Starting at = 0, outside air testing 005%, carbon dioxide is admitted atthe rate cof $000 cu ft/min. (a) What isthe percentage of carbon dioxide inthe ar in the room after 3 in? carbon dioxide? (©) When does the air in the room test 0.1%| | | | 2. 2B, 24, 25 33 mate'rrogiens 101, ‘The air ina room 50 ft by 20 ftby 8 ft tests 0.27 carbon dioxide. Startingat ¢ = 0, ‘outside air testing 0.05% carhon dioxide is admitted to the room. How many ‘cubic feet of this outside air must be admitted per minute in order that the air in the room test 0.1% atthe end of 30 min? Newton's law of cooling states that the rate at which a body cools proportional to the difference between the temperature of the body and that of the medium in which it is situated. A body of temperature 80 °F is placed at time 1=0 in a ‘medium the temperature of which is maintained at 50°F, At the end of S min, the body has cooled to a temperature of 70 °F. (2) What isthe temperature of the body at the end of 10 min? (b) When will the teroperature of the body be 60°F? ‘A body cools from 60 °C to $0°C in 15 min in air whieh is maintained to 30°C. How long will it take this body to cool from 100 °C to 80 °C in air that is main- {tained at 50 °C? Assume Newton's iaw of cooling (Exercise 23). ‘The rate at which a certain substance dissolves in water is proportional at the product of the amount undissolved and the difference e, — ¢,, where ¢, is the concentration in the saturated solution and ¢, isthe concentration in the actual solution. If saturated, $0 gm of water would dissolve 20 gm of the substance. IF 10 gm of the substance is placed in 50 gm of water and half of the substance is then dissolved in 90 min, how much will be dissolved in 3 hr?—— CHAPTER FOUR—— 44 Explicit Methods of Solving Higher-Order Linear Differential Equations ‘The subject of ordinary linear differential equations is one of great theoretical and practical importance. Theoretically, the subject is one of simplicity and elegance. Practically, linear differential equations originate in a variety of applications to science and engineering. Fortunately many of the linear differential equations that thus occur are of a special type, linear with constant coeficients, for which explicit methods of solution are available. The main purpose of this chapter is to study certain of these ‘methods. First, however, we need to consider certain basic theorems that will be used throughout the chapter. These theorems are stated and illustrated in Section 4.1, but proofs are omitted in thisintroductory section. By far the most important cases that of the second-order linear differential equation, and we shall explicitly consider and illustrate this case for each important concept and resnlt presented, In the Final section ‘of the chapter we return to this fundamental theory and present theorems and proofs in ‘this important special case. Proofs in the general case are given ia Chapter 11 BASIC THEORY OF LINEAR DIFFERENTIAL EQUATIONS. A. Definition and Basic Existencé Theorem DEFINITION A linear ordinary differential equation of ‘order n in the dependent variable y and the independent tariable x is an equation that is in; or canbe expressed in, the form ta Z+atdy= Fe, 4D41 BASIC THEORY OF LINEAR DifFERENTIAL EQUATIONS 103, sohere a snot identically zero, We shall assume that ay, ..,d, and F are continuous real functions ona real ntercala Example 4.1 The equation #y Be ‘sa linear ordinary ditferential equation of the second order. Y yee tarde cyee > example 4.2 ‘The equation ey. dy ae aE {is a'finear ordinary differential equation ofthe third order. dy a — sy asin + y ‘We now state the basic existence theorem for initial-value problems associated with and nth-order linear ordinary differential equation: ‘THEOREM 4.1 Hypothesis 1. Consider the nth-order linear differential equation anty 2 4 ay 0) 2 + aay = (4 als) Fy + lS +o Z + aye Fo) CN)104 PXPUCT METHODS OF SOLVING HIGHER ORDER UNEAR DAFERENTIAL EQUATIONS where doy... and F are continuous real functions on areal interval a Bample 43 ‘Consider the initial-value problem Py 5d ys, Grr wZrvyne, y(t) = 2, y@=—s, ‘The coefficients 1, 3x, and x?, as well as the nonhomogeneous term e*, in this second- order diflerential equation are all continuous forall values of x, ~00 < x < ao, The point x, here is the point 1, which certainly belongs to this interval; and the real numbers cq and c, are 2 and —5, respectively. Thus Theorem 4.1 assures us that a solution of the given problem exists is unique, and is defined for all x, —co < x < «@.41 BASIC THEORY OF LineaR DirreneNMiAL equations 105 D> Example'4.4 ‘Consider the initial-value problem ey ey ay 2S5exGr tie Psy 4) = 3, Y=, y@=-F Here we have a third-order problem, The coeflicients 2,x,3x7, and —5, as well as the nonhomogeneous term sin x, are all continuous for all x,—co Example 45 ‘The unique solution f of the third-order homogeneous equation @y pia Oro 28e p otha tyna,106 [EXPLICIT METHODS OF SOLVING HIGHER-OROER LINEAR OIFERENTIL EQUATIONS ‘which is such that JO) = $12) = fiy=0, is the trivial solution f such that (2) = O forall x. 8. The Homogeneous Equation ‘We now consider the fundamental results concerning the homogeneous equation arty we ‘We first state the following basic theorem: ey ay (0) $2 + asx) +4409 + ay =0. (42) THEOREM 4.2 BASIC THEOREM ON LINEAR HOMOGENEOUS: DIFFERENTIAL EQUATIONS Hypothesis. Let fu.fas---ufu be any m solutions of the homogeneous linear differ- ‘ential equation (4.2). Conclusion. Then cy fy +2 fot °*+ CwJae 18 also a solution of (4.2), where easly are m arbitrary constants Theorem 42 states that if m known solutions of (42) are each multiplied by an arbitrary constant and the resulting products are then added together, the resulting sums alo a solution of (42). Wemay pot this theorem ina very simple form by means of the concept of linear combination, which we now introduce. DEFINITION I SisSave-n fu ar@ m given functions, and ¢4,025---46q are m constants, then the x- pression ah teah te + eda is called a linear combination of f,sas--s Sux In terms ofthis concept, Theorem 4.2 may be stated as follows: THEOREM 4.2. (RESTATED) Any linear combination of solutions of the homogeneous linear differential equation (4.2) is also a solution of (4.2). In particular, any linear combination CA teaha tet ondeasic THEORY oF LINEAR onrexenriat equATions 107 of m solutions fiafz.---nfa of the second-order homogeneous linear differential equation 2 4 240) £3 + 04092 + ay00y =0 (4a) is also a solution of (44), > example 46 “The student will readily verily that sn x and cos x are solutions of 4, getyn ‘Theorem 42 states that the linear combination ¢, sin x + ¢; cos xisalso a solution for any constants c, and cs. For example, the particular linear combination Ssinx + 6008 x isa solution, D> sxample 4.7 ‘The student may verify that e, "4, and e* ae solutions of ay ty ay Gir eto. Theorem 42 states tha the linear combination ¢, e* + ¢,e"* + c3e**isalsoasolation for any constants cy, cand ¢. For example, the particular linear combination Det — 30° + Fe is a solution, We now consider what constitutes the so-called general solution of (42). To understand this we first introduce the concepts of linear dependence and linear independence. DEFINITION The m functions ffs fy are called linearly dependent on a x Example 4.8 ‘We observe that x and 2x are linearly dependent on the interval 0 x <1. For there ‘exist constants ¢, and ¢;, not both zero, uch that yx + (28) for all x on the interval 0 Example 4.9 We observe that sin x, 3sin x, and —sin x are linearly dependent on the interval 11s x:52.For there exist constants c,,c5.¢), nt all ro, such that ex sin x + 6,(3sin x) +¢4(~sin) = 0 for all x on the interval ~1 < x <2. For example let ¢, = I,¢a = 1ycy = 4. DEFINITION ‘Then functions sf.» f,are called linearly independent on the interval a < x < bif they are not linearly dependent there, That ithe functions fi fayn-f, ar linearly independent on a's x < bij the relation CAFO) + Oafabe) +--+ eh) = 0 forall x such that ax Example 4.11 We have observed that sin x and cos x ae solutions of ey grtys0 an for all x, ~20 < x < ep. Further, one can show that these two solutions are linearly independent. Now suppose f is any solution of (4.7). Then by Theorem 43 f can be expressed as a certain linear combination ¢, sin x +c, cos x of the two linearly independent solutions sin x and cos x by proper choice ofc, and ¢;. Thatis, there exist ‘wo particular constants ¢, and c, such that Sos) ey sinx + €3c08-x or)110 [EUCTT METHODS OF SOLVING HIGHER-ORDER LINEAR DIFFERENTIAL EQUATIONS forall x, ~c0 Example 4.12 We have observed that sin x and cos x are solutions of - {hy p00 forall x, ~20 < x < co, Firther, one ean show that these two solutions are linearly independent. Thus, they constitute a fundamental set of solutions of the given differential equation, and its general solution may be expressed as the linear combination €; sin x + ¢, 608%, where c, and cz are arbitrary constants. We write this as y = cy sin x + ¢, 605 x > Example 4.13 and e** of ®y fy ye $5-205 Ba ryao say be shown to belinearly independent for allx, — co < x < oo, Thus,e',e"% and constitute fundamental set of the given differential equation, and its general solution may be expressed as the linear combination ‘The solutions ee cet tee tee where ¢,,¢2, and ¢, are arbitrary constants. We write this as pager baet tose®. ‘The next theorem gives a simple criterion for determining whether or not n solutions of (4.2) are linearly independent. We first introduce another concept. DEFINITION Let fy Fanon fy be m real functions each of which has an (nm — {)st derivative on a reat interval a'< x beample 4.14 We apply Theotem 44 to show that the solutions sin x and cos x of dy aetyne441, SIC THEORY OF UNeAR DarERENTIAL EQUAMIONS 113 are linearly independent, We find that sinx cos. Wisin x,c08 x) = = Example 4.15 ‘The solutions e*, e€*, and ¢?* of ay a are linearly independent on every real interval for | jtoay Were eyelet et reac -1 |= 60 let e* ae ron al for all real x Exercises 1, Theorem 4.1 applics to one of the following problems but not to the other. Determine to which ofthe problems the theorem applis and state precisely the conclusion which can be drawn in this case. Explain why the theorem does not ‘apply tothe remaining problem. Py say 7 ‘ @ GrtsZtone O=8 yO=7. #y dy © Sesbrgne, yO=5 v0 2. Answer orally: Whi Tear Bertyno, 0 yo. he solution ofthe following initial-value problem? Why? 3. Prove Theorem 42 for the case m = = 2. That i, prove that if fu(x} and f(%) are two solutions of ay dy g(a) Fe + AULA) FE ala) y =O, then o, fe) + eft) is also a solution of this equation, where ¢, and cy are arbitrary constants114 cxruicrr merno0s oF soLvING HIGHER-ORDER LINEAR DIFFERENTIAL EQUATIONS. 4. Consider the atferential equation Py yay ae fag tO e { (a) Show that each of the functions * and eis a solution of differential equation (A) on the interval a < x

You might also like

• Fear: Trump in the White House

  

  From Everand

  Fear: Trump in the White House

  Bob Woodward

  Rating: 3.5 out of 5 stars

  3.5/5 (738)
• The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  

  From Everand

  The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  Viet Thanh Nguyen

  Rating: 4.5 out of 5 stars

  4.5/5 (122)
• A Man Called Ove: A Novel

  

  From Everand

  A Man Called Ove: A Novel

  Fredrik Backman

  Rating: 4.5 out of 5 stars

  4.5/5 (4610)
• • Magazines
  • Podcasts
  • Sheet music
• A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  

  From Everand

  A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  Dave Eggers

  Rating: 3.5 out of 5 stars

  3.5/5 (231)
• Grit: The Power of Passion and Perseverance

  

  From Everand

  Grit: The Power of Passion and Perseverance

  Angela Duckworth

  Rating: 4 out of 5 stars

  4/5 (589)
• The Little Book of Hygge: Danish Secrets to Happy Living

  

  From Everand

  The Little Book of Hygge: Danish Secrets to Happy Living

  Meik Wiking

  Rating: 3.5 out of 5 stars

  3.5/5 (401)
• Rise of ISIS: A Threat We Can't Ignore

  

  From Everand

  Rise of ISIS: A Threat We Can't Ignore

  Jay Sekulow

  Rating: 3.5 out of 5 stars

  3.5/5 (137)
• Principles: Life and Work

  

  From Everand

  Principles: Life and Work

  Ray Dalio

  Rating: 4 out of 5 stars

  4/5 (599)
• Shoe Dog: A Memoir by the Creator of Nike

  

  From Everand

  Shoe Dog: A Memoir by the Creator of Nike

  Phil Knight

  Rating: 4.5 out of 5 stars

  4.5/5 (537)
• Never Split the Difference: Negotiating As If Your Life Depended On It

  

  From Everand

  Never Split the Difference: Negotiating As If Your Life Depended On It

  Chris Voss

  Rating: 4.5 out of 5 stars

  4.5/5 (842)
• Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  

  From Everand

  Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  Margot Lee Shetterly

  Rating: 4 out of 5 stars

  4/5 (897)
• The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  

  From Everand

  The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  Mark Manson

  Rating: 4 out of 5 stars

  4/5 (5806)
• The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  

  From Everand

  The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  Ben Horowitz

  Rating: 4.5 out of 5 stars

  4.5/5 (345)
• Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  

  From Everand

  Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  Gilbert King

  Rating: 4.5 out of 5 stars

  4.5/5 (266)
• John Adams

  

  From Everand

  John Adams

  David McCullough

  Rating: 4.5 out of 5 stars

  4.5/5 (2409)
• The Emperor of All Maladies: A Biography of Cancer

  

  From Everand

  The Emperor of All Maladies: A Biography of Cancer

  Siddhartha Mukherjee

  Rating: 4.5 out of 5 stars

  4.5/5 (271)
• Team of Rivals: The Political Genius of Abraham Lincoln

  

  From Everand

  Team of Rivals: The Political Genius of Abraham Lincoln

  Doris Kearns Goodwin

  Rating: 4.5 out of 5 stars

  4.5/5 (234)
• The World Is Flat 3.0: A Brief History of the Twenty-first Century

  

  From Everand

  The World Is Flat 3.0: A Brief History of the Twenty-first Century

  Thomas L. Friedman

  Rating: 3.5 out of 5 stars

  3.5/5 (2259)
• Her Body and Other Parties: Stories

  

  From Everand

  Her Body and Other Parties: Stories

  Carmen Maria Machado

  Rating: 4 out of 5 stars

  4/5 (821)
• Yes Please

  

  From Everand

  Yes Please

  Amy Poehler

  Rating: 4 out of 5 stars

  4/5 (1898)
• The Glass Castle: A Memoir

  

  From Everand

  The Glass Castle: A Memoir

  Jeannette Walls

  Rating: 4.5 out of 5 stars

  4.5/5 (1715)
• Angela's Ashes: A Memoir

  

  From Everand

  Angela's Ashes: A Memoir

  Frank McCourt

  Rating: 4.5 out of 5 stars

  4.5/5 (440)
• The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  

  From Everand

  The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  Brene Brown

  Rating: 4 out of 5 stars

  4/5 (1091)
• The Light Between Oceans: A Novel

  

  From Everand

  The Light Between Oceans: A Novel

  M.L. Stedman

  Rating: 4.5 out of 5 stars

  4.5/5 (789)
• A Tree Grows in Brooklyn

  

  From Everand

  A Tree Grows in Brooklyn

  Betty Smith

  Rating: 4.5 out of 5 stars

  4.5/5 (1929)
• The Art of Racing in the Rain: A Novel

  

  From Everand

  The Art of Racing in the Rain: A Novel

  Garth Stein

  Rating: 4 out of 5 stars

  4/5 (4203)
• The Perks of Being a Wallflower

  

  From Everand

  The Perks of Being a Wallflower

  Stephen Chbosky

  Rating: 4.5 out of 5 stars

  4.5/5 (2104)
• Wolf Hall: A Novel

  

  From Everand

  Wolf Hall: A Novel

  Hilary Mantel

  Rating: 4 out of 5 stars

  4/5 (3811)
• The Woman in Cabin 10

  

  From Everand

  The Woman in Cabin 10

  Ruth Ware

  Rating: 3.5 out of 5 stars

  3.5/5 (2520)
• Sing, Unburied, Sing: A Novel

  

  From Everand

  Sing, Unburied, Sing: A Novel

  Jesmyn Ward

  Rating: 4 out of 5 stars

  4/5 (1104)
• Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  

  From Everand

  Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  Ashlee Vance

  Rating: 4.5 out of 5 stars

  4.5/5 (474)
• On Fire: The (Burning) Case for a Green New Deal

  

  From Everand

  On Fire: The (Burning) Case for a Green New Deal

  Naomi Klein

  Rating: 4 out of 5 stars

  4/5 (74)
• The Outsider: A Novel

  

  From Everand

  The Outsider: A Novel

  Stephen King

  Rating: 4 out of 5 stars

  4/5 (1840)
• The Yellow House: A Memoir (2019 National Book Award Winner)

  

  From Everand

  The Yellow House: A Memoir (2019 National Book Award Winner)

  Sarah M. Broom

  Rating: 4 out of 5 stars

  4/5 (98)
• The Unwinding: An Inner History of the New America

  

  From Everand

  The Unwinding: An Inner History of the New America

  George Packer

  Rating: 4 out of 5 stars

  4/5 (45)
• The Constant Gardener: A Novel

  

  From Everand

  The Constant Gardener: A Novel

  John le Carré

  Rating: 3.5 out of 5 stars

  3.5/5 (104)
• Brooklyn: A Novel

  

  From Everand

  Brooklyn: A Novel

  Colm Tóibín

  Rating: 3.5 out of 5 stars

  3.5/5 (1946)
• Little Women

  

  From Everand

  Little Women

  Louisa May Alcott

  Rating: 4 out of 5 stars

  4/5 (104)
• Steve Jobs

  

  From Everand

  Steve Jobs

  Walter Isaacson

  Rating: 4.5 out of 5 stars

  4.5/5 (806)
• Manhattan Beach: A Novel

  

  From Everand

  Manhattan Beach: A Novel

  Jennifer Egan

  Rating: 3.5 out of 5 stars

  3.5/5 (792)
• Bad Feminist: Essays

  

  From Everand

  Bad Feminist: Essays

  Roxane Gay

  Rating: 4 out of 5 stars

  4/5 (1016)
• ASTM

  

  Document14 pages

  ASTM

  Anonymous cG96JOJtl

  No ratings yet
• 001 Art-Appreciation

  

  Document13 pages

  001 Art-Appreciation

  Btrz Qnns

  No ratings yet
• Isit328 2019 PPT 04 Cryptography

  

  Document100 pages

  Isit328 2019 PPT 04 Cryptography

  SHA Eric

  No ratings yet
• TCPIP Protocol Suite

  

  Document27 pages

  TCPIP Protocol Suite

  Vedant Aggrawal

  No ratings yet
• BS 4999-143-1987 Requirements For Rotating Electrical Machin

  

  Document8 pages

  BS 4999-143-1987 Requirements For Rotating Electrical Machin

  James Alejo

  No ratings yet
• Ag 180ao MMM 000

  

  Document294 pages

  Ag 180ao MMM 000

  sergey62

  No ratings yet
• Design of Bolted Connections

  

  Document7 pages

  Design of Bolted Connections

  Engineer

  No ratings yet
• Belt and Pulley Systems

  

  Document5 pages

  Belt and Pulley Systems

  NBRellos

  50% (2)
• District Reading Monitoring and Evaluation

  

  Document8 pages

  District Reading Monitoring and Evaluation

  Merafe Ebreo Aluan

  No ratings yet
• Archmodels 06

  

  Document5 pages

  Archmodels 06

  JEAN CARLO LARA PEREZ

  No ratings yet
• Nuclear Physics Checklist

  

  Document16 pages

  Nuclear Physics Checklist

  George Chahni

  No ratings yet
• Oracle Mid Test

  

  Document17 pages

  Oracle Mid Test

  Alin Ionut

  No ratings yet
• Giancoli - Physics Principles Appendix-2

  

  Document1 page

  Giancoli - Physics Principles Appendix-2

  Aman Kelta

  No ratings yet
• Laplace

  

  Document33 pages

  Laplace

  sr samie

  No ratings yet
• Business Research Methods: Session 6 To 7

  

  Document109 pages

  Business Research Methods: Session 6 To 7

  Contact Info

  No ratings yet
• 2018 Math Quiz

  

  Document15 pages

  2018 Math Quiz

  maria arliza torio

  No ratings yet
• Concepts in Balanced Device Measurements

  

  Document12 pages

  Concepts in Balanced Device Measurements

  A. Villa

  No ratings yet
• Vector Worksheet 10 Ans

  

  Document13 pages

  Vector Worksheet 10 Ans

  Megan Goh

  No ratings yet
• CATALOGO PRODOTTI 2013 Per Sito WEB TED

  

  Document52 pages

  CATALOGO PRODOTTI 2013 Per Sito WEB TED

  muhammad abidin

  100% (1)
• Chapter - 3 Information Security

  

  Document94 pages

  Chapter - 3 Information Security

  Jebesa

  No ratings yet
• Effects of Irregularities of Electric Charges

  

  Document2 pages

  Effects of Irregularities of Electric Charges

  Rodel Espartinas Elicot

  No ratings yet
• Finite Element Analysis of Hospital BED

  

  Document10 pages

  Finite Element Analysis of Hospital BED

  G SATHVIK

  No ratings yet
• T G 1653646560 Time Zones Powerpoint Ver 1

  

  Document8 pages

  T G 1653646560 Time Zones Powerpoint Ver 1

  Malvika Balchandani

  No ratings yet
• Internship Report

  

  Document29 pages

  Internship Report

  Pooja kore

  No ratings yet
• HMT SK

  

  Document60 pages

  HMT SK

  sunilkewat11390

  No ratings yet
• Transformer Life Extention Methodlogy Thru LTC Replacement

  

  Document28 pages

  Transformer Life Extention Methodlogy Thru LTC Replacement

  lbk50

  No ratings yet
• Module 4: Plant Growth Factors and Latest Trend in Crop Science

  

  Document26 pages

  Module 4: Plant Growth Factors and Latest Trend in Crop Science

  Norhanah Dionisio Baruba

  No ratings yet
• Chapters 29 and 35 Thermochemistry and Chemical Thermodynamics

  

  Document54 pages

  Chapters 29 and 35 Thermochemistry and Chemical Thermodynamics

  Danice Amie Concepcion

  No ratings yet
• Cinema - Problem Description

  

  Document10 pages

  Cinema - Problem Description

  Vlada

  No ratings yet