INSTRUCTOR’S SOLUTIONS MANUAL MARK WOODARD Furman University CALCULUS AND CALCULUS EARLY TRANSCENDENTALS William Briggs University of Colorado at Denver Lyle Cochran Whitworth University Bernard Gillett University of Colorado at Boulder with the assistance of Eric Schulz Walla Walla Community College Boston Columbus Indianapolis New York San Francisco Upper Saddle River Amsterdam Cape Town Dubai London Madrid Milan Munich Paris Montreal Toronto Delhi Mexico City Sao Paulo Sydney Hong Kong Seoul Singapore Taipei TokyoThis workis protected by United States copyright laws andis provided solely for the use of instructors in teaching their courses and assessing student learning. Dissemination or sale of any part of this work (including on the World Wide Web) will destroy the intearity of the work and is not permi ted. The work and materials from it should never be made available to students except by instructors using the accompanying text in thei classes. All recipients of this work are expected to abide by these restrictions and to honor the intended pedagogical purposes and the needs of other instructors who rely on these materials. The author and publisher of this book have used their best efforts in preparing this book. These efforts include the development, research, and testing of the theories and programs to determine their effectiveness. The author and. publisher make no Warranty of any kind, expressed or implied, with regard to these programs or the documentation contained in this book. The author and publisher shall not be liable in any event for incidental or consequential damages in connection with, or arising out of, the furnishing, performance, or use of these programs. Reproduced by Pearson from electronic files supplied by the author. Copyright © 2011 Pearson Education, Inc. Publishing as Pearson, 75 Arlington Street, Boston, MA 02116. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written, permission of the publisher. Printed in the United States of America www.pearsonhighered.comContents DI: Differential Equations D1L1 Basic Ideas D1.2 Direction Fields and Euler's Method D1.3 Separable Differential Equations D1.4 Special First-Order Linear Differential Equations D1.5 Modeling with Differential Equations D1.6 Chapter Review 20 30 4CONTENTS Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-WesleyDifferential Equations D1.1_ Basic Ideas D1.1.1 Second-order, because the highest-order derivative appearing in the equation is second order. D1.1.2 Linear, because the unknown function and its derivatives appear only to the first power D1.1.3 The equation is second-order, so we expect two arbitrary constants in the general solution, D114 We have y(0) =C +10 5, so C= —5. The solution is y(t) = ~5e-™ + 10, D1.1.5 Yes. Note that y'(t) = 0 and y/(t) = 2. D1.1.6 No. (0) =6 43. D1.1.7 Yes, it is a solution. Note that y/(t) = -5Ce~™, so y(t) + 5y(t) = 0. D1.1.8 Yes, it isa solution. y'(0) 301-4, so ty'(t) + 8y(t) = 301 +301 = 0, D1.1.9 Yes, it is a solution. y'(t) = 4C) cost — ACysindt, so y"(t) = —16Cy sin tt — 16C2 cos 4t, so y"(t) + 1y(t) = 0. D1.1.10 Yes, it isa solution. y'(r) = —Cye* + Cze*, s0 y"(x) = Cre + Cxe*, $0 y"(x) ~ yl) = 0. D1.L.11 Yes, it is a solution. y'(t) = 32e2*, so y/(t) — 2y(t) = 32% — (82% — 20) = 20. Also, y(0) = 6 10=6. D1.1.12 Yes, it is a solution. y'(t) = 480°, so ty'(t) ~ 6y(t) = 48t" —48¢° + 18 = 18. Also, y(1) = D1.1.13 Yes, it isa solution. y/(t) = 9sin 3, so y"(t) = 27eos 3¢. Thu 0. Also, y/(0) = 0 and y(0) = y"(t)+9y(t) = 27 eos 3t—27 cos 3t = DL.L.14 Yes, itis a solution, y/(2) = 4(2e2* + 2e-%) and y""(x) = $(4e2* — de®). So y"(2) — dy(x) = 0. Also, y(0) = 0 and y'(0) = 1. DLL18 y(t) = fB+e™) dt = Bt eo" 4. DL.1.16 y(t) = f(12" — 2004 + 26-2) dt = 2 — 40 20+ E+. DL.1.17 y(2) = f(dtan 2x — eos) dx = ~2n |cos22|~3sinze + C = 2In|see2x| —3sine + C. D1.1.18 plz) = f(G2-® — 5 + M2) de = —2e-* —5r +227 +0. D1.1.19 y/(t) = f(60e! — 4+ 124-8) de = 1245 — at — Gi? +E. y(t) = S28 — At 6? + C)at = 2° — 20 + 61 Cit + Co D11.20 y'(t) = f(i5e + sindt) dt = e — Leosdt + C1. ylt) = f(Ge" — beosdt + Cy) dt = Se" — Sysindt + Cit + Ca4 CONTENTS: 240, +58} + Cr+ Cy DLL.21 ul(2) = (5509 + 3627 — 2129 + 102-8) de = 5c + Ba8 — (2) = f(6.52"° + 328 — $a — 52-2 + C) de = fa! + BaP 4 DL1.22 v'(2) = feet de re® — De" + Cir + Ce, ee HCY. Ufa) Jive" — e+ Cy)de ne ECHO D11.23 y(t) = +C, we have C= 3. Thus, y(t) =t-+et+3. (I+el)dt=t+e'+C. Because y(0) = D1.1.24 y(t) = f(sint + 00821) dt Thus, y(t) = —edst-+ }sin 2¢-+5. -1+C, we have C cost + $sin2t + C. Because (0 D1.1.25 yl) = f(30? — 3x~)de = 2 42-8 + C. Because y(1) = 0 =1+1+4C, we have C = 2. So ple) = 23 a8 = 2, D1.1.26 ye) = 0+ C, we have C = 8 Thus le) = 2tan e+ JfAsec?2rde = 2tan2e +C. Because y(0) 8 0+ Ci, we have Cr = 01 DLL.27 y(t) = f(t — Be) dt = 6 — St + Cy. Because y'(0) = 0 1, Thus, y(t) = 20-041 ult) = J(GE2 504) dt = 2°°-P-+C2, Because y(0) = 1 = 0-0+Cz, we have Cz D1.1.28 u'(x) = (de® —8e-™) de u(z) = f(2e* +de-™ —3) de = Thus, u(x) = e* — 2e-2* — Br +2. 20? + 4e* 4-C, Because u'(0) “23240, Because (0) = 1 24+44+Ch, we have C; 1-2-0+C2, we have C: D1.1.29 a. v(t) = -9.8t-+ 20.4. s(t) = 4.90 + 29.48 + 30, b, The object reaches its high point when ~9.8¢ + 29.4 = 0, or t= 32 (3) = TALL meters, 3. At that time its position is D1.1.30 a. v(t) = -9.8¢-+49. s(t) = 4.97 + 49¢ + 60. b. The object reaches its high point when —9.8¢-+49 = 0, or t= 8% = 5. At that time its position is (5) = 182.5 meters D1.1.31 We have p(t) = (1500-2011 }e+20H. The amount of resource is increasing when 1500—20H > 0. which occurs for H < 75. The amount of resource is constant when 1500—20H = 0, which occurs for H = 7. If H = 100, the resource is zero when (1500 — 2000}e% + 2000 = 0, which oceurs for t = 201n4 a: 28, D1.1.32 We have p(t) = (po10000}e®%" 410000. The amount of resource is deereasing when po—10000 < 0. ‘oF py < 10,000. The amount of resource is constant when py = 10,000. If py = 9000, the resource vanishes when —1000e% = ~10000, or t = 20!n 10 a 46, piss ‘The height function is given by AQ) = (Vimo 242.4)" (4-047. Therm | is empty when h(t) 0, which occurs after about 5.6 seconds 4 Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.D1. BASIC IDEAS 5 Dis ‘The height function is given by (0) = 4 (v225 ~ XFS 5)" (1.5 ~.5580)2. The tank a iS empty when (0) =0, which occurs after about 4 2.71 seconds 4 D1.L.as 1a. False. That is a specific solution. The general solution is ¢ + C. b. False, It is second order, but is not linear. c. True. First find the general solution, and then find the specific solution which satisfies the initial condition, D1.1.36 y(t) = f(tnt + 1)dt= t+ ftlntdt. Let w= Int and do = t, so that du= # and v= 2/2. Then ut) = t+ (@nt)/2— ft/2dt = t+ (@int)/2- 2/44. DLLAT ule) = f Bey de~ f aby de = ine? +4) —tan"(e/2) +0. D1.1.88 Note that pty = 2p — gly. Thus, o(t) =f what = J (; D11.39 y/(2) = f opr de. Let w= 1 — 2°, so that du = ~2rde. Substituting gives ula) = 3B fw 8? du = uP + Cy = he + Crd. ylo +01) de = sin“! (w) + Cyr + Cp. D1.1.40 Let w te! —e +C. Because y(0) and dy = edt, Then du = dt and v 1=0-1+C, we have C 0-0+C, we have C= 2. DLLAL u(r J (ste -4) da = }tan™"(x/4) —4a+C. Because u(0) Thus, u(2) = } tan“"(x/4) — de +2, D1.1.42 plr) = f 2 ae =I(2 - +) de = 2In| Fy +€. Because p(1) = C= =210(/2) = 21n2, Ths, p(a) = 204 (oey| +22 2in(1/2) +0, we have D1.1.48 Using the result of number 40 above, we have y/(t) = tet—e+Cy, and becanse y/(0) we have C; = 2. Thus y!(t) = tet —e' +2. y(t) = fya te! — Yet +2t + Cy. Because y(0) ~2404Cs, we bi 150-14, Sitet — 8 +2) dt = tet ee $y = re Cy = 2. Thus, y(t) = tet — Qe! +26 +2. D1.1.44 u(t) = Ce) 1-9 = AO. Thus wi(t) + 48 = HO + 4 <0, DLLAS w(t) = Chet + Cae! +Cate!, and w(t) = Cre! + Cae! + Cae! + Cate! = Chet + 20 re! + Cate! Thus, u(t) — Qu'(t) + u(t) = (Che! + 2C ze + Cate!) — Cre! + Cae! + Cate!) + Cyet + Cate! = 0. DLLAG g'(2) = -20ye°% + Cre? + -20 26-2, 0 gf"(2) = 4Cye~* —20€~-* + ~ 2090-9" AC are = AC e-™* 40 3¢° 24 4Cpre-™, Thus, ”(2)+-49"(1) +49(2) = AC e2* 4090-44020 +4( 200-2 Cre-® + ~2Cpxe-™) + (Cre + Core™ $2) = 8. DLALAT w(t) =2C,t + 8C2t?, so w"(t) = 2Cy + 6Cat. Thus, Pull(t) — Atul(t) + Gu(t) = 20,0? + 6Cxt3 — 420, + 30218) + GCL? + 6C2t* = 0. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.6 CONTENTS: D118 u(t) = 5Cyt! — ACat-$ — 312, so w"(t) = 2006 + 200-8 — Gt. Thus, u(t) — ult) = 20,6 + 20C2t~+ — Ge — 20 (Cyt? + Cat * = 88) = ae D119 2"(t) = —Cyent + 20y¢% — 3Cye-M — ef. So 2M(t) = Cyert + AC ae + 9Cge—M — et and (1) = Cre + 802e% — 27050 —e, Thus 2(0) + 22M (t) — 52!(t) — G2(t) = ~Crert + 8Cae* — 2700 — FC et + 8x0 + 18C pe + 5Cyef — 100 ze + 1500 + Set = 60 — 6026" — 600° + Ge! = Bet D1.1.50 a. y(t) = Chet ~ Cye~*, s0 y(t) = Cret + Cae. Thus, y!"(t) — y(t) = 0. b. w(t) = 2Cre% — 2020, so y(t) = ACae* + 4Cae-™*, Thus, y(t) — Ay(t) = 0. c. Tt appears that a general solution should be CyeM + Cze*. ‘Then y/(t) = kCie* — kCxe~™, and y"(t) = RCye + Exe. Thus, y(t) — y(t) = 0. 4. If y(t) = Cy cosh kt + Casinh kt, then y(t) = kCisinh kt + C2 cosh kt and y(t) = KC) cosh kt + F2Cy sinh kt. Thus y(t) — y(t) = 0. DLLs a. y(t) = Creost — Crsint, so y"(t) sin — Czeost. Thus, y(t) + (0 2C2.cos 2t ~ 2Cy sin 2t, so y/"(t) = ~4C sin 2t — 4Cy cos 24. Thus, y"(2) + Ay(t) = 0. c. A general solution appears to be y(t) = Cisin kt + Ca cos kt. Then y'(t) = kC1 coskt ~ kC2 sin kt, so y"(0) = ~K2Cy sin kt ~ k?C2 cos kt. And then y""(t) + y(t) = D152 a, Let m(t) = £(1 eM), Note that m(0) = 0. Then m'(#) = £(ke~™). Therefore, ty eho = Be) + et) aren ted b. We have m(t) = 200(1 ost) . Tt appears that lim, 5 m(t) = 200. D1.L.s3 a, Let p(t) = Note that 1 £ = 1 pb = Se). We have yay Kore! cont WO= Tecemp Teer Troe P(t Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.DL. BASIC IDEAS b. IF p(0) 50 = oH, then 50-+50C = K, so C= Ke © We have p(t) =~. 4. Tint, ue Fear = $2 = 300, which is consistent with the graph from part c DLL.54 a. Let v(t) = $(1—e™). Then v(0) = 0, and = bet = ge b. With b= 0.1, we have o(t) = 98(1 ~~), Him pgs ot) = 98. D1.1.55 a. IE y(t) = woe, then y(0) = yo, and y(t) = —kuve™*, 80 v(t) = kyle) b. Let y(t) = ger. Then y(0) = yo, and y'(Q) = gate = —A(u())* c. The first order reaction decays more quickly. D1.1.56 a. Let M(t) = K (42). Note that In(M(t)/K) = er n(Mo/K) ( Mit) = K (48) In(M/K) (re) = Aey* M(t) n(M(t)/). Also, M(O) = K(Mo/K) = Mo. Copyright ©2011 Pearson Education, Ine. Publishing as Addison-Wesley8 CONTENTS Ing K = 200, My = 100, and r = .05, we have M(t) = (48)"" = 2001/2)" ©. Lim yop M(t) = 200 = K. D1.2 Direction Fields and Euler’s Method D1.2.1 Choose a regular grid of points in the ty-plane, and for each point P, make a small line segment with slope f(t,y). D1.2.2 It will have slope 32 — 3(1)? = 6. D1.2. tw = y(8) = 1. uy = up + f(3,1)(1) = 1+.6= 1.6, D1.2.4 Because the differential equation is giving the slope at a given point, we can approximate the solution to the differential equation by starting at the initial point, and using the slope to guide where the next iteration should be. In essence, we are numerically "following the direction field” to estimate the solution to the differential equation. D1.2.5 D126 DL27 fa, This matches with D. b, This matches with B. ¢. This matches with A. 4. This matches with C. Copyright ©2011 Pearson Education, Ine. Publishing as Addison-Wesley1.2. DIRECTION FIELDS AND EULER'S METHOD. D1.2.8 Note that the slopes are zero when y = —1 and when t = 1. Also, they are positive when both, y>—Land (> 1, The only differential equation with this property is choice a, D129 An initial condition of y(0) = —1 leads to a constant solution. For any other initial con- dition, the solutions are increasing over time, D12.10 An initial condition of y(0) = 1 leads to 4 constant solution. For any other initial condition, the solutions oscillate between in- creasing and decreasing over intervals of time length one, DL211 An initial condition of y(0) = 1 leads to a constant solution, Initial conditions y(0) = A lead to solutions that are increasing over time if A > 1 and solutions that are decreasing over time if A < 1. “TSNN3-2772-NNS. ANSE S SRRSSNS SSN SS SSNS Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.10 CONTENTS VOLE VLCC VELOCE, VOC, ean SOO G7 OLS S SSSSASASRANS SAAS D217 a. The solutions y= 1 and y = —1 are constant. b. Solutions are increasing when both y > 1 and y > 1, (so for y > 1) and when both y <1 and y <-1 (s0 for y < -1). In other words, for |y| > 1. Solutions are decreasing for |y| <1 ¢. Initial condition y(0) = A leads to solutions that are increasing over time if [Al > 1 and decreasing over time if [Al < 1 DL2.18 and y = —1 are constant. fa. The solutions y = b, Solutions are increasing when both y > 2 and y > 1, (so for y > 2) or when both y <2 and y <1 (s0 for y < 1). Solutions are decreasing for =< y <2, Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.1.2. DIRECTION FIELDS AND EULER'S METHOD. u nT A ae VE Sie ¢. Initial condition y(0) = A leads to solutions RQYQAHHHY that ave increasing over tine i> 2 or Ae SATAY 1, and decreasing over time if —1 < A < 2. SASSAAS ASA AAS SINANRAAARAANAN: RRR I D12.19 a. The solutions y= 7/2 and y = —n/2 are constant. b. Solutions are increasing when |y| < 7/2 and decreasing when 1/2 < |y| <7. ¢. Initial condition y(0) = a leads to solutions that are increasing over time if |A| < x/2 and decreasing over time if |A| > 7/2. D1.2.20 a. The solutions y = 0, y = 3. wid y = 4 are constant. b. Solutions are increa and when y > 4. ing when y < ~3 and when 0 < y <4. Solutions are decreasing when ~3 < y <0 . Initial condition y(0) = A leads to solutions that are increasing over time if A < ~ 0 4, Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.2 CONTENTS D1.2.21 Equilibrium solutions are P(t) = 0 and P(t) = 500, D1.2.22 Equilibrium solutions are P(é) = 0 and P(t) = 1200. D1.2.23 VSASASgyggyes SNS i = i Equilibrium solutions are P(t) = 0 and P(t) = OG, 3200. (00704 Wat 4404 : We Ws LEE Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.1.2. DIRECTION FIELDS AND EULER'S METHOD. 13 D1.2.2 4 Equilibrium solutions are P(t) = 0 and P(t) = 50. D1.2.25 up = 2 wy = 2+ f(0,2)(.5) = 2+ (A)(5) =A. uy = A+ f(5,4)(.5) = 4+ (8)(.5) = 8. So y(.5) 24 ) and y(1) =:8. DL2.26 up = 1. uy = 1+ f(0,-1)(2)= = 64. So y(.2) x —8 and y( 4) =~ 64 1+1(2) = —.8. ug = 8+ f(2,-8)(2) = - 8+ (8)(.2) = D1.2.27 up = 1. wy = 14 (0,14) = 1D LA uy = 11+ f(A, LA)(A) = 1.1 + (9)(4) = 1.49. So y() = 11 and y(.2) = 1.19, D1.2.28 up = 4. uy = 4+ f(0,4)(.5) = 4+ (A)(5) = 6. uy = 6 + J(.5,6)(.5) = 6 + (6.5)(.5) = 9.25. So y(.5) a2 6 and y(1) = 9.25, D1.2.29 approximation of (0.2) | approximation of y(0.4) 8 64 ala 81 65610 05 SUM51 (66342 025, 81665 (66692 0.818731 and et = 67032, ‘eqror in approximation of y(0.2) | eror in approximation of y(0.4) uss 03082 n OO87S, 1 05 00122, (00600) 0% (00208) (00340) c. The time step At = 025 has the smallest errors. A smaller time step generally produces more accurate results. 4. Halving the time steps results in approximately halving the error D1.2.30 At_| approximation of y(0.2) [ approximation of y(0-4) 2 242 ala 243101 05 243681 0% 2.43087 Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-Wesleyu 21034 and 2e m2 2.44281 At_| exvor in approximation of y(0. 2) [error in approximation of y(0.4) 2 0.0103418 0.022805 1 (0.00534184 0.011793) 05 0.00271605, 0.005900 025, 0.00136003) 0.00302622) c. The time step At =.025 has the smallest errors. A smaller time step generally produces more accurate results. 4. Halving the time steps results in approximately halving the error D1.2.31 At [approximation of y(0.2) | approximation of y(0) 2 3.2 3.36 ala 3.19 3.3139 05 3.18540 3.33058 025, 3.18395 3.38308 b. 4 e7 a 3.18127 and 4 ~ et = 3.92068, At_[ ervor in approximation of (0.2) | eror in approximation of (04) 2 0.0187308 0.03032 1 (0.00873075, 0.01422, 05 (0.0042245 0.00680961 025, 0.0020789 (0.00330088) ¢. The time step At = .025 has the smallest errors. results. 2 A smaller time step generally produces more accurate Halving the time steps results in approximately halving the error ar approxtim tion of y(0.2) approximation of y(0.4) 2 2 18 ala 52 05 23 5a 025, 235 55 b. 224+.2= 24 and 42+ 4 = 56, At_[ ervor in approximation of y(0.2) | eror in approximation of (0.4) 2 0.04 0.08 1 0.02 0.04 05 0.01 0.02, 025, 0.005 0.01 . The time step At = 025 has the smallest errors. results. A smaller time step generally produces more accurate Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-WesleyDS AND EULER'S METHOD 15 1.2. DIRECTION F 4d. Halving the time steps results in approximately halving the error D1.2.33 a. The computations yield t [o[o2[o4 | 06 [ os L 12 14 16 Ls. 2 vw. | 1 [0.6 [0.36 [0.216 [0.1206 [0.07776 [ 0.046656 [ 0.0270036 [0.0167962 [0.010077 | o.00604662 So y(2) = 0.00604602. 00183156, so the error is about 0.0183156 — 0.00604662 = 0.012269, . The computations yield: 09) 1 0.134218 tu [ofor[o2 [os [oa [05 06 ux [1 [08 [0.64 [os12 [0.4006 | o.s276s [0.202144 ies coreor [ooiaat [oss | wf [ie [is [ia [is 7 6 ux [0.0859 | 0.0687 [ 0.0550 | 0.0440 [0.0352 [0.0! So 9(2) = OOT15S The enor is about 0 O1SI56-— 0.01155 56 piss a The computations vil clolelor[o pos [1 [| a weft foo [se [136 [asi [aossoe [raises [2sss0s apie [os mp Le pos ps] vx [uae [zosoo [zorser [Roasan [zonoo [Roorrs [aanast [zooei® | So y(3) = 2.99812. b. (3) = (8 —4e~®) a 2.99008, so the error is about 2.99812 ~ 2.99008 = 0.00804, . The computations yield: ipo[o [pos [om [os [oe [or [os [oe Do vu [tf 02 [oe [oon [rsnre] iosnas [Losi [oot [saxon [aaasis | aoTs tle Le ps De pos poe Da pe | ws [2560 | 2512 [rani [DaDwos [Dasoos [aeerar [Doone [Roeros [Dowzan [DonHRS | tLe p22 pos | 2s 2s | 26 m= [2 3] vn [oes [2oroay [aoress [aos [aos [aosror [aos [200220 [2 vossr [2050s | 2 9505 = 0.00807 So y(3) % 2.99505 The error is about 2.99812 4d. The error with twice as many steps is less than half the other error D1.2.35 a. After many calculations, we arrive at y(4) a2 3.05765. Copyright © 2011 Pearson Edueation, Inc. Publishing as Addison-Wesley6 CONTENTS b. y(d) =3 + 5e~* = 3.00158, so the error is about 3.09158 — 3.05765 = 03398. . After many calculations, we arrive at y(4) = 8.0739, The ervor is about 3.09158 ~ 3.0739 = 0.01768. 4. The error with twice as many steps is about half the other error D1.2.36 45125. a. After many caleulations, we arrive at y(2) b. y(d) = VB = 44721 so the error is about 4AT214 ~ 4.45125 — 02089 ¢. After many calculations, we artive at y(2) = 446178. ‘The error is about 447214 ~ 4.46173 = 0.01041 4. The error with twice as many steps is about half the other err pi2.37 a. Tne. b. False, It allows you to compute approximations, D1.2.38 a. y = 2 is an equilibrium solution, because 2(-2) +4 =0. bande. D1.2.39 a. y = 3 is an equilibrium solution, because 6 ~ 2(3) =0. bande. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.1.2. DIRECTION FIELDS AND EULER'S METHOD. a D1.2.40 a, Solve y(2—y) = 0 to get equilibrium solutions y=Oand y= bande. D1.2.41 {XSAN .0..8.8 NANSANSSSSSSSNN NNN NNNNNNNN a. Solve y(y—3) = 0 to get equilibrium solutions y=Oand y= bande. D1.2.42 a. Solve siny = 0 to get equilibrium solutions y= kr, where k is any integer bande. ~ Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.18 CONTENTS D1.2.43 a. The equilib and y =3. jum solutions are y = 0, y = band ¢ 1.2.44 a. Solving y’ = 0 gives the equilibrium solution y = ~b/a, which is a horizontal line. Note that the general solutions y = (A+ # decreases without bound if a > 0 but A < = 4 increnses without bound if a > 0 and A> But if a <0, the general solutions have limit rust increase to it if A < D1.2.45, a Ar be b. Recall that up = A and f =a. Sous = A+ f(a, A) (52) Cte = ue + F(t. ue) (Bet). where ty = a+ & (22) for k= 0.1. Copyright ©2011 Pearson Education, Ine. Publishing as Addison-Wesley19 1.2. DIRECTION FIELDS AND EULER'S METHOD. D12.46 b. The equilibrium solution is m(t) = 200. . The solutions are increasing for A < 200 and decreasing for A > 200, D247 b. The solutions are increasing for A < 98 and decreasing for A> 98, 98. cc. The equilibrium solution is w(t) D12.48 a. In both cases, the equilibrium is 0. as Addison-Wesley. Copyright ©2011 Pearson Education, Ine. Publishin,20 CONTENTS: b. The second order approaches more quickly. D1.2.49 a. We have twp = y(0) = 1, and user = ug + f(testtn)h = ue + auch = up(1 + ah) for k= 0,1,2 (1+ ah)*(1 +ah) = (1+ an)§, & 0.1.2. b. Suppose ux = (1-+ah)*. Then uie = ue(1-+ah) im, go(1 + ah)!*/* = lim, 49 ((1 + ah)!/*)"* = (e#)! ce. limy.yo te = lms yo(1 + ah) ent = ylte). D12.50 a. We have tuy = y(0) = 1, and user = ue + f(testte)h = ue — auch = up(1 ah) for k= 0,1,2 b. Suppose uy = (1—ah)*. Then uee1 = ue(1— ah) = (1 ah)*(1— ah) = (1 ah), k= 0.12, c. The function r* increases as = increases exactly when |r| > 1 and decreases exactly when |r| < 1. So ug = (1—ah)* will inerease when k increases exactly when |1—ah| > and will decrease as kx increases exactly when [1 — ah| <1. 4. Suppose [1 ~ ah <1. Then -1 <1 ah < 1. Subtracting one from everything gives -2 < ~ah <0. Dividing through by —a (which isa negative ammber) gives 0 < ft 2/a, as desired D1.3 Separable Differential Equations DL.3.1 A separable first-order differential equation is one that can be written in the form g(y)y'(#) = h(t), where the factor g(y) isa function of y and h(t) is a function of t D1.3.2 Yes, this equation is separable because it can be written in the form y2y'(t) = (-2(t-+ 4) DL.3.8 No, this equation cannot be written in the required form, D1.3.4 Integrate both sides with respect to and convert the integral on the left side to an integral with, respect to y. DLB.5 We have y'(t) ® 30 [Bat = [Pat soy =HA4C. D136 y!(t) =5e~, so fy (thdt = [ bed, so y = Be" +. DIST yi =802, 50 fydy = fede. Thus, =e +C, and thas y = 4VIFFT DILL8.8 We have £84 = 22 +1, s0 f 8dr = f(e? +1) de, Thus, Iny = 23/8-+2+C, and y= Acl#"/9+2 DL.3.9 We have fe~"/? dy = fsintdt, and so ~2e-¥/? = —cost-+C. Thus, y = ~2ln (4 cost + C) D1.3.10 We have [wo'/?dw = f hdr = f (2+ 4) de, so Qw'/? inel|- b +0)" inja| — 2 +C, and thas w = Dis. LY = 4.0 f Pde = f de. so J +C arbitrary Constant C' by its opposite, this ean be written as y = ye ‘Thus, y = If we replace the Dis.az pr) = pc Tle, f py = Satay dts We hve SiG? +4) = Steir +6 so In(y? +4) = Cy ~ greta. This can be written as y? +4 = Aen H)"*/2 “obl= VAEC Ea D1.8.18 SPH =sint, so J SPA ae = fsintat. Thus, 2 = cost +C. Solving for y gives y= +b Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.13. SEPARABLE DIFFERENTIAL EQUATIONS, a DLS. Boyt) =e, 50 f ag dy = fede, so 4 tan*(y/2) Cy — 40, and y = 2tan(Cy — de"), 24.0, Ths, tan(y/2) = D1LB.18 (2) = 2% 90 fetdu= J 2dr and e* = fet +0, Thus, w= In (3 +0) Ain |r| +. +o DL.3.17 This is separable, and can be written as y'(t) =}. Thus, fy'(@®dt = f 4 =Int+C, so ylt) = Int +C. Because y(1) +C, we have C = 2. Thus, y(t) = Int +2. 1.8.18 This is separable, and can be written as y/(t) = cost. Thus y(t) 1=0+C, we have C= 1, Thus, y(t) =sint +1 sin +, and because y(0) = 1D1.3.19 This is separable, and is already written in the desited form. We have f2ydy = [32 dt, so y= 1+. Because y(0) = 9, we have 81 = C, so y = VEST. Dis .20 This equation is not separable. D1.3.21 This equation is not separable, D1.8.22 This equation is separable. We have f= f(4¢ +1) at, and thus In|y|=t* +t+C. Therefore, eel! et Ae“*, Substituting y(0) =4 gives A = 4, so the solution to this initial value problem is nae D1.3.28 This equation is separable. We have f 2ydy = fetdt, so y? = et +C, and thus y = +V FC. Substituting y(In2) = 1 gives 1 = 2+ C so @ = —1, and the solution to this initial value problem is vera .B.24 This equation is separable. We have fy-dy = feosrde, so -¥2 = sine +C. ‘Therefore 4(—2sinx + C)¥/2, Substituting y(0) = 3 gives C = 1/9, so the solution to this initial value problem is y = (-2sinx +3) D1.3.25 This equation is separable. We have fe¥ dy = fe* dx, and thus e = & +C. Therefore, y = In(e* + C). Substituting y(0) = In3 gives In3 = In(1 + C), so C= 2 and the solution to this initial value problem is y= In(e* +2) D1.3.26 This equation is separable. We have [ have 1=1+€, $0 C =0. Thus, y= tan <2 ydy = J dt, so tany =£+C. Because y(1) = »/4 we DL3.27 yl) =t s0 fydy = Jtat, s0 7/2= 2/240. Because y(1) = 2, we have 2 = 1/2+C, s0 0 = 3/2. Soy = 243. ‘The solution corresponds to the upper portion of the curve, Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.CONTENTS (2- yy) = 142,50 f(2—y)dy = f+) de and 2y~ y?/2= 2+ 27/2+C. Because y(1) = 1 we have 2~1/2=1+1/2+C, so C =0. Thus, 2y — 42/2 = 2 + 22/2 describes the solution. D1.3.29 oe. We rel nace Loren O jO fcos(/2) dx. We have —cosu wwe ulm) = 7/2, we have —cosa/2 = +€, $00 = (1) +€, so C = ~2. Thus, ~cosu = 2sin(r/2) ~ 2 describes the solution. - ‘The curve in the middle in the rightmost column, of curves is the particular one described by our D1.3.30 (2+ oP uyl(a) = 2x, 50 f(2+ v2Pyy(a) de = JPorde, and thus 2" = 2240. Because y) = -1, we have @ = 2. Thus, 2+ 92)* = Ga? 2, and thas y? = YO FAT —2 oop Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-WesleyD13. SEPARABLE DIFFERENTIAL EQUATIONS 23 D1.3.31 Voriy(2) = ETT, so fygrady = [VEFTae, so 2y +4)? = Fe +1240. Because y(3) = 5, we have 2(27) = 48+, so C=. Thus, 3(y +4)? = He +1)? +8, s0 (y +4)? = (@+1)¥? +19 D1.3.82 set 0) = hp 90 fof = f oy. Ths }tan7 (2/2) = btan7(2/4) +. Because 2(4) = 2, we have m/8 = x/16+C, so C = x/16. Thus, 2tan“(2/2 (o/4) + n/4 describes the so- Iution, D1.3.33 a. This equation is separable, so we have fey dP = 008d, 0 L(+ abr) dP = v.08 + 0. Therefore Infasi-p| = 0.08¢ + C. Substituting P(0) = 50 gives —In3 = C, and solving for P gives Pi) = gare b. The steady-state population is ims. P(t) = 200. D1.3.34 a. This equation is separable, so we have Sopher = frat, 0 f (p+ dP = At+D. Therefore, In|p| = kt + D, which is equivalent to <2, — CeM, Substituting P(0) = Pa gives C= Pa/(A— Ph), and solving for P gives P(t) = pace Copyright ©2011 Pearson Education, Ine Publishing as Addison-Wesley24 CONTENTS c. The denominator in P(t) above is positive for all t > 0 when 0 < Py < A, so P(t) is defined for all £ > 0: we have lim 5 P(t) = A, which is the steady-state solution, 3.35 a, True. It ean be written as ua!(x) = b. False, c. True. When separated, we have ye%y'(x) = x, and the left-hand side of the equation can be integrated by parts. D1.3.86 We have fevy/(t)de = f ¥ dt, oe” = #24 C. Because y(1) = Ind, we have 2 = 0+ C, so of = +2, Ths, y= In (1 DL8.87 We have rut) = 4.50 f aay dy = f Bat, and thus In| weer as pa| = 8th |t1+C. Because y(1) 1 we have C = —In2. Thus, in| 27] = $in t|—In2 describes the solution. Then 22, = #8/2,s01— 25 = 18/2 so phy = 258. Thus y +1 = 2p, 80 = ah D1.3.38 f2ydy = feos? tat, so y? = § + 342 +C. Because y(0) = -2, we have 4 =0+0-+C, s0 C= 4. Thus, y= —y/p + B+. ‘D1.3.39 Assume y > —3 and t > ~6/5. J (sa) wae = ff pAgdt, so In(y +3) = 249 +0, We can write this as 5 In(y +3) = In(5t +6) +5C, so (y +3)° = A(6t +6). Because y(2) = 0, we have 3° = 16A, so A=3/16, We have (y +3) = SE(t-+6)". Thus, y = 8+ S92 61 +0)! = 8-2-4964 50)!" D1s.s0 a. fy? dy = J(P +2t/3)dt, so & = 2/3 + 2/3 +C. It is convenient to write this as y? = # + + Cy b. When y(=1) = 1, we have 1 = -1+1+4 Ci, so Cy = 1. When y(1) = 0, we have 0 =1+1+ Ch, so (Cy ==. When y(=1) = ~1, we have -1 = -1 +14), 80 C= 4 a: DLs4L a. few dy = f(Arsine?—2) de. Thus, — —2eos:2?—29/2+K., Then e~¥/? = cosa?+22/44C, $0 —y/2 = In(cos.2? + 2?/4+C), and y = ~2In(cos.2? + 22/4 +0). b. When y(0) = 0 we have 0 = -2In(1 + C), so @ = 0__When y(0) = In(1/4), we have In(1/4) = =2In(1 + C), so In2 = In(1 +C), so C= 1. When y( 772) = 0. we have 0 = —2In(0 + 4/8 + C, so c=1- Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.D13. SEPARABLE DIFFERENTIAL EQUATIONS 1.3.42 a. de + yh b. Curves are orthogonal when their slopes are negative reciprocals of each other, and the negative reciprocal of =2! is 2. ©. We have 28 = have y? = ke so 2In|y| = ny? = In|2| +C. Thus, y? = e@|2|. We can write +e = k, so we 1.3.43 Differentiating implicitly gives 2° + 2yy/ = 0. s0 yf = 42 ¥. We have #2 = #2, so in|y| = In|2| +C so y iS the collection of Curves y = kz We are thus secking curves so that |x| = ke. So the family of curves we are seeking D1.3.44 a, This equation is sepatable, so we have f sbzydy = fat, so f ($+ 74;) dy = kt-+ D. Therefore wf = kt-+D, which is equivalent to -2; = Ce, Substituting y(0) = yo gives C = yo/(1 — wo), and this y = qtr “ ¢. The denominator in y(t) above is positive for all £ > 0 when 0 < y < 1, so y(t) is defined for all t > 0; we have lity 405 y(t) = 1, which is the steady-state solution D1.3.45 a, We have mu"(t) = mg ~ kv2, so u'(t) = g — av? with a = k/m. b. We solve av? = g to obtain the terminal velocity 6 = 97a = y/gm]F. Copyright ©2011 Pearson Education, Ine. Publishing as Addison-WesleyCONTENTS . This equation is separable, so we have dy=t+D. hs, she f (abs ~ ce) = + D, and gy In 4. Ce Phe initial condition (0) ~ 0 gives C = —1, and solving for v gives » = 3522.0, This ean be written as v= (/a7a) See a, We have a = 0.1, 09.90 m/s ] D1.3.46 a. We have mo'(t) = mg ~ Rv, so v(t) = g bv with b= R/m. b. Solve bu =g to obtaia terminal velocity # = g/b = mg/R. €. The equation v" = g— bv is first-order linear, with general solution » = Ce~® 4 8. The initial condition (0) =0 gives C = —8, which gives v = a(1—e-®). , We have b= 0.1, 7 = 98 m/s. D1.3.47 a. The equation h = 2kVF is separable, so we have f st = f kat, so Vi = kt-+C. The initial condition 10) = H gives C= VIF, so the solution is h = (VHT-+ kt) b. The solution for k= —0.1 and M = 0.5 is h = (0.7071 — 0.102. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.D13. SEPARABLE DIFFERENTIAL EQUATIONS a ¢. The tank is drained when A(t) = 0, which gives ¢ = —VH/k. With k = —.1 and H = 5, we have t= f5/.1 = 7.07 seconds. a Diss a. The general solution to y? = ky is y = cot b. The equation y/ = —ky? is separable, so we os. have =f = fhadt, so = kt +C. The initial condition y(0) = yp gives C = 1/4, tnd solving for y sis y= pene = eae : : pis.a9 1, The growth rate is positive when 0< M <4. ‘The fimetion R(M) has derivative R'(M) = =r (in (#2) +M- 2) = —r (In (48) +1) which is 0 when M/4 = Ife or MI =4/e. We also observe that limyy yo» R(M) = 0 and RA) = 0, so R(M) takes its maximum at the evitieal point MI = 4Ye. sparable, so we have f arabes = frdt, so In|lnM — In K| = —rt-+ D, and “rt, Therefore M = K. b. The equation is thn In (28) = C ‘The conditions r = 1, K = 4 and Mo = 1 give C = ~Ind and M = 4!" = 4, Observe that lim 99 M(t) = de” = 4, so the limiting size of the tumor is 4 c. In general, the limiting size of the tumor is limy-4, Ke” = Ki, because r > 0. D1.3.50 Solving with a CAS gives, Copyright ©2011 Pearson Education, Ine. Publishing as Addison-WesleyCONTENTS DLs.51 a. yl) =1,s0 f= fidt, Thus =! =¢+C. Because yy = 1 we have -1=C. Thus, y= 74 b. Bu'(t) =1,s0 f= fd. Thus gb =t+C. Because yo = Jy we have -1 = C. Thus and y = hey u'(t) = 1, s0 J fr = ftdt. Thus = t+C. Because yo = n~'/" we have —1 = C. Thus, phe = 1 tand ny” = phy. Thus, y= coger We have an, 1 gage = 00 D1.s.52 a. tay (t) = aebay: 80S arty = Stay. Using partial fraction decompositions, we have In | 2%] = in| |+K. Thus |2 Dye Thus, 1 hy = £Dy/Fh, and thus 1 = Dl = Fy. 80 y+l= oe, and y= = AP VEE. cis can be written as 22VEE sR. Be = TY iby yee vet ae corn where C= Sy b. Ify(1) = A, then A © WA=1, then C= 3, so the solution is y= = 2. Tinsel) = hy * 6.46 Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-WesleyD13. SEPARABLE DIFFERENTIAL EQUATIONS 29 e. When y(1) = 2, we have © = 3, s0 the solution is y= xX — flim. yg y(t) = he = 16 via Tava 8 Time goo y(t) = ch D1.3.53 a. Luy'(t)dt = fle /2+t)dt, 90 y2/2=e/2+2/2+C,s0y =e +2 + Cy, Soy=tVEFETO. b. Iy(-1) =1, then 1= YTEFIFE 80 C1 = ~1/e, s0y = y If y(—1) =2, then 2= YT/EF ITE 0 C, = 3 1/e,s0 y= YO FESS Te. c. For > 0, the solutions increase as f increases. -VIJeF TFG, 0 Cy = -1/e and y = — YF FP HT. -VIJeF TFG 80 C, = 3- Ie, soy = - VF FEFE— Te. c. For t > 0, the solutions decrease as t increases, a. If y(—1) = ~1, then -1 If y(-1) = 2, then -2 Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-Wesley30 CONTENTS: D1.4 Special First-Order Linear Differential Equations D1.4.1 Because y(0) = 4, we have 4 = C~ 13, so C = 17. Thus, the solution is y= 17e—2" — 13. D1.4.2 The general solution is y = Ce% — $12 = Ce™ +4 D1.4.3 The general solution is y= C Fee 8 D1.4.4 The equilibrium solution is y= 3. It is unstable. D145 Because k= 3 and b= —4, we have y= Ce" $ D1.4.6 Because k = —1 and b=2, we have y= Ce* +2, D1.A4.7 Because k = —2 and b Cot — 2, 4, we have y D1.4.8 Because k= 2 and b= 6, we have y= C2* ~ 3. D1.4.9 Because k= 12 and b= 15, we have w= Ce“ + § 1.4.10 Because k= $ and b= 14, we have v = Cev/2 — 28, D1.4.11 Because & = 3 and b= 6, we have y = Ce +2. Because y(0) = 9, we have 9 =C+2, s0 Thus, the solution is y = 7e™ +2 D1.4.12 Because k C= ~4. Thus, the solution is y Land b = 2, we have y= Ce~* +2. Because y(0) = —2, we have -2 = C +2, so a4eF +2, D1.4.18 Because k Thus, the solution is y Ce — 4, Because y(0) 2 and b= 8, we have y ea D1.4.14 Because k = 2 and b = 6, we have u = Ce* — 3, Because u(1) = 6, we have 6 = Ce? ~ 3 C= 9e-®. Thus, the solution is u = 92"? — 3, D1.4.15 Because k = 3 and b = 12, we have y = Ce —4, Because y(1) = 4, we have 4 = Ce — 4, so C= 8e-®. Thus, the solution is y= 8" — 4, D1.4.16 Because k = —1/2 and so C= ~12e~"/?, ‘Thus, the solution is 2 = — 6, we have z = Ce~!/2+12, Because 2(—1) = 0, we have 0 = Ce? +12, HH/2-/2 4 19 D417 ‘The equilibrium solution is y solution is unstable. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.Dia, SPECIAL FIRST-ORDER LINEAR DIFFERENTIAL EQUATIONS D14.18 ‘The equilibrium solution is y = 2. The solution stable. D14.19 ‘The equilibrium solution is y = —3. The solution is stable. D1.4.20 ‘The equilibrium solution is y = —4, The solution is unstable. y erent] FREER RSS SO ES Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.32 CONTENTS D421 ‘The equilibrium solution is w= —3. The solution is stable. D1.4.22 1 eee ‘The equilibrium solution is u = 8/4. The sohi- x tion is unstable. pias Because k = .005 and b = —500, we have B = “y Ce 5 588 — Ce-M5t.+ 199000. Because B(O) = 50,000, we have 50000 = C + 100000, so C = met —50000. Thus, B = —50000e-%°* + 100000. ent The balance is zero when t = In(2)/.005 = 139m months, Pepenpenpenenntent pia2a Because k = (01 and 6 = ~750, we have B= a Ce + 0 — Cel + 75000. Because B(0) = cal 415,000, we have 4500) = C+ 75000, 80 C = 30000. Thus, B = —30000e-°!" + 75000. at The balance is zero when t= In(5/2)/.01 292 months : Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.Dia, SPECIAL FIRST-ORDER LINEAR DIFFERENTIAL EQUATIONS 33 1.4.25 Because k 0075 and b 1500, we have ame B= Cet + 59 = Cet + 200000. Because som B(0) = 100,000, we have 100000 = C + 200000, ool so C = 100000. Thus, B = 100000675 + ‘eal 200000 =| The balance is zero when £ = In(2)/.0075 = 93 months. a 1.4.26 Because k = .004 and b = —800, we have B = ~ Cet 4. 800, = Ce MOA + 999000, Because B(0) = aol 40,000, we have 40000 = C’ + 200000, so C= af 160000, ‘Thus, B = ~160000e-%* + 200000, ‘The balance is zero when t = In(5/4)/.004 == 56 vo months. Laapeeeapennnpegeeeng #495, Because G5e~ 9S" +95. We have 1In(13)/.08 = 32 minutes D1.4.27 We know thi 85, we have 65e~* + 25 30 when 65e~% +25 after the coffee is first poured. the solution has the form T(t) = (90 ~ 25)e7M + 25 = 85. Thus, k 08. Thus, T(2) 30, oF when e~°8 = 2. This occurs when ¢ D1.4.28 We have T(t) = (900 ~ 30)e~-"2" +30, We have T(t) = 100 when 870e~%% +30 = 100, which 2 = 2. This oceurs when ¢ = 82/52 ws 136 ‘occurs when e~ D1.4.29 T(t) = (5 ~ 20)e- +20 In(15/13) © 143. Thus T(t) = —15e which occurs when e~!8" = 2/15, Thus, when t = MH 4.20. Because T(1) = 7, we have —15e~ +20 = 7, so 94's 20. ‘The milk, will reach 18 degrees when —15e~ 420 = 18, 1n(2/is) 1d minutes. 1.4.30 T(t) = (100 — 10)e~# + 10 = 90e-™ + 10. We are given that T(30) = 80, so 80 = 90e~"* + 10, so ¢-8 = 7/9, Thus k = !2242) se 0084. The soup will reach 30 degrees when 30 = 90e~-"S# +10, which ‘occurs when e084 = 9/9, or t= 22/9) 170 minutes Dish a. False, Ib is y(t) = Ce** +9 where C is an arbitrary constant b. True, Note that if y(f) =0, then y/(¢) = not K(0) — b= —bas dictated by the equation. c. False, It is not separable, 4. False. It approaches it asymptotically. 1.4.82 First note that if ky-+b < 0, then [Ay-+b| =—ky—D. We will use this fact in what follows. We have ris '(t) = 1,80 f ly = dt, so £n(—ky~b) = t+Cs. Thus hn(—ky—b) = Kt-+C2, so —hy b= Cae" and y= Ce — 8 D14.33 f(ty'(t) + ylt)) dt = (1 + tat, so ty(t) = t+ 2/2+C. Thus, y(t) = 141/2+C/t. Because y(l) =4, we have 4=1+1/2+C, so C= 5/2. Thhs, y(t) = 1+ t/2+5/(21) ay D1.4.34 f(ty!(t)+3t?y(t)) dt = f(1/t+1) dt. Thus, ty(t) = in|t]+t+C. Therefore, y ue) &. tole Because y(1) = 6, we have 6 = 0-+1+C, so C= 5. The solution is thus y(t) = "+ b+ & Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.34 CONTENTS D1A.35 f(ety!(t) — e~tylt)) dt = fe dt, so e~*y(t) = 2/2 +.C. Thus, y(t) = e/2+ Cet. Because (0) =4, we have 4 = 1/2+C, so C= 7/2. The solution is therefore y(t) = e/2 + Tet/2. D1.4.36 f(t? + 1)y'(t) + 2ty(t)) dt = f 30? de, so (1? + y(t) = 0 +C. Thus, y(t) = py + p&p. Because: 4(2) = 8, we have 8 = 8/5 + C/5, so C = 82. The solution is therefore y(t) = #232 D187 a. Note that k = .03 and 6 = ~600. Thus, B(t) = Ce +20, 000. If By = B(0) is the amount borrowed, then By = C+ 20,000, so C = By ~ 20,000 = 20, 000 ‘Thus, B(t) = 20,0006 + 20,000. This is an increasing fimetion because its derivative is positive. It is occurring because the amount paid monthly is less than the monthly accruing interest b. Because 20000 - 03 = 600, the amount borrowed should be less than 20000 if the balance is to be decreasing, ©. The maximum amount that can be borrowed By — m=0, or By = D1.4.88 a. T(t) = (Ty ~ A)e"# + AL We are seeking thy» 80 that T(ty2) (Tq — A)e“*72, 50 nd not have the unpaid balance increase is By when J. ‘Thus we have To/2— A = b. As k increases, thy» decreases ©. If A > Tp/2, then the equation Ty/2— A = (Ty ~ A)e~'# would have no solution, because the left-hand side is negative and the right-hand side is positive. D14.39 0.02 5 150, fa. The general solution is y(#) = Ce substitute y(0) = 0 to obtain C= -150; hence y(t) = 1501 — b. The steady-state level is limy_44g 150(1~ e-®") = 150mg, €. We have 150(1 ~ e-P%) = 0,9-150, so e0% = 0,1, and thus ¢= #242 =. 115.1 hours. D1.4.40 a. The general solution is y(t) = Ce" + 10); substitute y(0) = 500 to obtain C + 100 = 500, so C= 500 — 10d; hence y(t) = (500 — 10b)e%°" + 108. . In this ease the solution is y(t) = 100e! + “ 400, which approaches 90 as t + o0. vo Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.Dla, SPECIAL FIRST-ORDER LINEAR DIFFERENTIAL EQUATIONS 35 In this case the solution is y(t) = —100e" + al 600, which reaches 0 when ¢°°! = 6 or t = snl 100 In6 = 179 yrs, oa Di441 a. The equation y/(t) = 0.008y—h has steady-state solution y = h/0.008 = 125h, so we solve yo = 2000 = 125h to obtain h = 16. b. If A = 200, then the steady-state solution is y = 125 - 200 = 25,000, 1.4.42 with general solution B = Ce" +m/r; in this ease r = 5000, so B = 20,000—5000° 9%, fa. The equation B = rB ~ mis first-order lin 0.05, m/r = 20,000, and the initial condition By = 15,000 gives C ‘The balance decreases, "e) solution is B = m/r = $50,000, which gives m = 0.05 - 50,000 = b. The steady-state (constant ba $2500. 14.43 a. Let v= y!®. Then v(t) = (1 —p)y*y'(t), 80 v'() = -1/(f) + ay! =b, so b. Given y(t) + ay = by®, we have {u"(t) + ay = by, so w() = (p— 1av(t) +41 ~p). Then v(t) = Cele" — Msp) = Cele! 4. 8. Therefore, ue wo (cern 2 D144 1, Using the results of the previous problem, we have a = 1, b= 2, and p= 2. Thus ult) = (Ce +2)* and b= 3, so wo =( c. We have a=1, b= 1, and p= 1/2, s0 b. We have a = —2, p= ule) = (Cet? 1 Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.36 CONTENTS: D1.4.45 Let p(t) = exp(f1/tat) = e!* = t, The original differential equation can be written as t(y'(0) + (1/0\y(t)) = 0, or ty/(t) + ylt) =O. Integrating both sides with respect to ¢ gives ty(t) = C, 30 yl) = Because y(1) = 6, we have C = Gand y = D1.4.46 Let p(t) = exp([3/tdt) = e¥™* = 3, The original differential equation can be written as °(y/(0) + (3/t)y(t)) = 8 — 2t*, or ty'(t) + 3t2y(t) = t — 2¢", Integrating both sides with respect to t gives y(t) = 1/420 /5-+C, s0 ult) = t/4—22/5-+C/t®, Because y(2) = 0, we have 0= 4 $+, and solving for C gives C= 4, Thus, y(t) abt a DL4AT Let plt) = exp(f pdt) = explln(t? +1) = +1. The original differential equation can be written as (#2 + 1)y'(t) + (2t)y(t) = (2 + 1)(1 + 342). Integrating both sides with respect to t gives (2+ Vy(t) = f(t! +42 +1) dt = 3/5 +40°/3-+t+C. Because y(1) = 4, we have 8 = 3/5 +4/3+14C, so C= B. Ths, yt) — MERI a azo asc ‘D1.4.48 Let p(t) = exp(f2tdt) =e. The original differential equation can be written as ef y'(t) + ef (21)y(t) = 3te. Integrating both sides with respect to t gives ey(t) = $e + C. Because y(0) = 1, we have 1= $+C, 90 C= 5}. Thus, u(t) D1.5 Modeling with Differential Equations D1.5.1 The growth rate function specifies the rate of growth of the population. If the growth rate function is, positive, then the population is increasing, while the population is decreasing when the growth rate funetion is negative. D1.5.2 The carrying capacity is the upper limit of the size of a population, due to limitations in resources. Mathematically, it appears as a horizontal asymptote as f — 90. D1.5.3 If the growth rate function is positive and decreasing, then the population is increasing. Whether or not the population is increasing is completely determined by whether the growth rate function is positive or negative, D1.5.4 A stirred tank reaction takes place in a tank that is filled with a soluble substance like salt or sugar ‘The tank has an inflow and an outflow pipe, so the volume of the solution is constant. The tank is assumed to be completely stirred at all times, so the solution is uniform. The problem is to find the mass of the substance in the tank at all times DL.5.5 Is is a linear, first-order differential equation. DLs. # In the absence of prey, the predator population deceases exponentially, while encounters between prey ‘and predators increase the predator population (the prey are the food supply), ‘* In the absence of predators, the prey population increases exponentially, while encounters between the prey and predators deplete the prey population (the predators eat the prey). DL.5.7 The solution curves in the FH-plane are closed curves that circulate around the equilibrium point. D1.5.8 They both oscillate cyclically, with the prey population peaking slightly before the predator popu- lation, Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.DL 5. MODELING WITH DIFFERENTIAL EQUATIONS at D1. D1. D1. D1. 5.9 D15.10 5.1L D15.12 5.18 Dis.14 5.15 We must soke PQ) = 2P(1-£). We have [20 dt = f.2dt, which ean be writ ven as f(Eil + gla) dt = f2de Ths tn (gill) = 26+ ©. Taking the exponential of both sides and reciprocating gives MPO. — Ae Because P(0) = 50, we have A= 5. Thus 2 = Be +1, $0 Plt) = Ber. 5.16 We must solve P(t) = 4P(1— xq). We have f pam dt = fAdt, which can be writ- ven as (5 In (saptnag) = 4t+C. Taking the exponential of ) ae = fade, Tw both sides and reciprocating gives S774. — Because P(0) = 100, we have A = 54. Thus Bde! + 1, 80 P(t) = 5 Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-Wesley pita38 CONTENTS DL.s.17 We must solve P(Q) = rP(1 fs). We have f PR dt = frat, which can be write ten as (FE + sett) le = frat, Thus In (aati) = rt-+C. Taking the exponential of both sides and reciprocating gives Wwo—PH) get war Because P(0) = 200, we have A= 9, Thus #82 = ert 1, so P(t) = 200%... Now because P(1) = 600, we have r = In(27/7). So P(t) = sone. D1.5.18 We must solve P(t) = rP(1~ da). We have f ten as [ (Sip + rom) dt = frat. Thos, tn (rela) = 14+C. Taking the exponential of both = frdt, which can be writ. sides and reciprocating gives 22" — Aett, Be cause P(0) = 50, wehave A= 2, Thus #2 = 26-41 so P(t) = zi... Now because P(1) = 60, we have 1 = In(4/3). 80 P(t) = samt rc D1.5.19 We have AO =—r, 90 f atl dt = f —rdt. Note that In(M/K) = (K/M)(1/K)M"() = M’(®)/M. ‘Thus integrating gives ln |Iu(M/K)] = —rt-+ €. and thus In(M/K) = Ae. Because M(0) = Mo, we have A = In(Mo/K}) Thus M(t) = K exp(tu(Mo/K)e")) = K (42)" D1.5.20 With r=.1, 100, and My = 50, we have 5 ay" suo =10() D1.5.21 With r =.05, K = 1200, and My = 90, we have By" ane) =120 (8) Copyright ©2011 Pearson Education, Ine. Publishing as Addison-WesleyWITH DIFFERENTIAL EQUATIONS 39 D1L5. MODELIN D1.5.22 With » = 6, 1 = 5500, and My = 20, we have M(t) = 5500 (ge) D1.5.23 mass of copper sulfate is mp = 0. We have a. The int ~ giant) + 20-4 = aabmn(t) +80. mi(t) b, This is an equation of the form m'(t) = km +b, so the solution is m(t) = Ce! — 8 = Ce~*" + 10,000. Because m(0) = 0, we have C = —10,000, Thus, ‘m(t) = —10, 000e-"°8 + 10, 000. D1.5.24 a. The initial mass of the salt is mp = 3000. We have m!(t) = —q2gi(t) + 20-3 = shm(t) + 60. b, This is an equation of the form m'(t) = km +, so the solution is m(t) = Cel — B= Ce" + 30,000, Because m(0) = 3000, we have C = ~27,000. ‘Thus, ‘m(t) = —27, 000e-""" + 30, 000. D1.5.25 a, The initial mass of the sugar is my = 2000 - 40 = 80,000. We have m/(t) = —zitgnn() +1010 = Fan(t) + 100, b. ‘This is an equation of the form m'(t) = km +b, so the solution is m(t) = Cet — £ = Ce~ + 20,000, Because m(0) = $0000, we have C = 60,000. Thus, m(t) = 60, 000e% + 20, 000. D15.26 ass of the pollutant is my = 1000000 20 = 20,000,000. We have mé(#) = — 72282 m(t) + m(t). fa. The initial 0-120 = Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.40 CONTENTS b. ‘This is an equation of the form m'(t) = km +b, s0 ot the solution is m(t) = Cet ~ # = Ce~12*, Because na ‘m(0) = 20,000,000, we have C= 20,000,000. Thus, oo (t) = 20,000, 000e~ 2. ‘The pond will have a mass oa of 10 percent of the initial value when .1 = e~ 4, so ol # = —In(.1)/.0012 = 1919 hours. ow D15.27 a. xis the predator and y is the prey b. ~8r + Gry = 0 when 2 = 0 or y= y— Ary =0 when y= 0 or x= 1/4. ‘The desired lines are ¢ = 1/4 and y= 1/2. c. The equilibrium points are where both equations are zero simultaneously, so they are (0,0) and (1/4,1/2) 4. Note that o”(t) = 32(2y —1) and y(t) = y(t — 42). For 0 <2 <1/dand 0 0. For 0.<2< 1/4 and y > 1/2, we have 2’ > O and y! > 0. For x > 1/4 and 0 1/4 and y > 1/2, we have 2! > 0 and y! <0. . The direction of the solution is clockwise. ain QE, D15.28 a. xis the prey and y is the predator b. Qe dey = 0 when ¢ =0 or y= 1/2. -y + 2ry = 0 when y= 0 or x= 1/2, ‘The desired lines are ¢ = 1/2 and y= 1/2. c. The equilibrium points are where both equations are zero simultaneously, so they are (0,0) and (1/2,1/2) 4. Note that o”(t) = 2x(1— 2y) and y(t) = yx 1). ~~ For 0. 0 and y! <0, = For 0 << 1/2and y > 1/2, we have 2 <0 and y' <0. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.1.5. MODELING WITH DIFFERENTIAL EQUATIONS: aL = For 2 > 1/2 and 0 0 and y' > 0. ~ For x > 1/2 and y > 1/2, we have 2" <0 and y’ > 0. . The direction of the solution is counterclockwise, D1.5.29 a. 2 is the predator and y is the prey. b. ~8r-+2ry =0 when 2 =0 or y=3. 2y~2y=0 when y= 0 or = ‘The desired lines are «= 2 and y= 3. c. The equilibrium points are where both equations are zero simultaneously, so they are (0,0) and (2,3) a. Note that 2/(t) = a(y— 3) and y'(#) = y2~2). For 0<2<2and 0 0. For 0< <2 and y > 3, we have 2! > 0 and yf > 0. For x > 2and 0 2and y > 3, we have 2! > O and y’ <0. . The direction of the solution is clockwise. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.2 CONTENTS: D15.30 a. xis the prey and y is the predator b. 22 —2y =Owhen 2 =0 or y= 2. —y-+2ry =0 wh y=Oore ‘The desired lines are «= 1 and y= 2. c. The equilibrium points are where both equations are zero simultaneously, so they are (0,0) and (1,2) a. Note that 2”(t) = (2 y) and y'(t) = y(x— 1), For 0<2<1and 0 O and y' <0. For 0<2<1and y > 2, we have 2! 1and 0 0 and y/ > 0. For £> Land y > 2, we have 2! <0 and y > 0. . The direction of the solution is counterclockwise, DL5.a1 a, True, The growth rate function is the derivative, so where itis positive, the population is inereasing b, Thue, In the limit, the solution in the tank is the same as the solution being poured in ¢. True. In the absence of predators, we assume that the prey population increases exponentially. D1s5.32 a, f(P) = r(1~ 2) + rP( (122). ‘This is zero when P = %, and an analysis of the sign of f/ shows that f' > 0 for P< Sand f' <0 when P > . ‘Thus there is a maximum of F(K/2) = r(K/2)(1/2) = rK/4 at P= K/2. b. J(AM) = ~rtn(M/K) ~ pM(K/M\(A/K) = —r(ln( M/K) + 1). This is zero when In(M/K) = 1, ot MM/K =e, which oceuss for M =. Note that f(K/e) = —r(K/e) n(1/e) = rK/e. Au analysis of the sign of f* shows that J’ > 0 for Af < and f? <0 when M >. ‘Thus there is a maximum at M Kye. 1.5.33 We must solve P(t) = rP(1—£). We have f KG dt = frdt, which can be written as I (pe + hn) dt = frdt, Thus, kn (x2htn) = rt+C. Taking the exponential of both sides and reciprocating gives SR = Ae-*t, Because P(0) = Ph, we have A so P(t) KP AGP. Thus py = (Abbe +1 Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.D1L5. MODELIN WITH DIFFERENTIAL EQUATIONS 43 Y= (Ay D15.34 M(0) = K (Me ® K =K. = My. Beeanse lim, yg eT = 0, we have lim sgg M(t) = a. Note that this differential equation is first-order linear with k = =# and b = C;R, so the solution is m(t) = C&PIME GE CeRIVIES CLV. Because m(0) = mo, we have mo = C+ CV, so C= mo ~C,V. Thus, the solution is m(t) = (mo — C:V}e-®"Y + GV. b. m(0) = (no ~ CV) + CV = mo. ©. Note that lime jg, e“/Y = 0, so lim gom(t) = C:V. In the limit, the solution in the tank is the incoming solution, so the amount of material in the tank is the amount per unit volume times the volume of the tank . Increasing R causes the graph to approach the asymptote more quickly. D15.36 The initial mass of the drug in the body is 0. We have m/(#) = —%m(t)-+(.06)(500) = —015m(¢) +30, ‘This is an equation of the form m'(t) = km +b, s0 of the solution is m(t) = Cel — £ = Ce~915" + 2000. b. Because m{(0) = 0, we have C ="-2000. Thus, m(t) = “| 2000!" + 2000. | c. Because tinny -s9g ¢5! = 0, limi say m(t) = 2000. a. The drug mass reaches 1800 mg when 1800 = —2000e-°!% + 2000, so when 1 = eso t = = In(.1)/.015 «2 153.5 minutes. DL5.a7 a, We can write the equation as 1"(t) = h(t). This isa first-order linear equation with k = 3 b=0. Thus, the solution is 1(@) = Ce! — 2 = AcE, Because 1(0) = %, we have A=}. Thus, 1(0) = e-W/*C, The eurvent decays exponentially with decay constant 2. onder linear equation with k = 7b gente 4 VC, Baste Ce RE +-VC. In the long run, the charge has b. We can write the equation as Q'(t) = 7hQ +H. This is a . Thus, the solution is Q(t) = Cel ~ £ = Ae-t/C Q(0) = 0, we have A= -VC. Thus, Q(t) = (-' and b= limit VC) D15.38 a. P(t) = Pie’ = 28le". Because P(10) = 281e!" = 310, we have r = In(310/281)/10 = .0098. b. TEPC) = op becae = peepee and if P(50) = 489, then 2a = 439. Solving for KC gives K = 3903 ¢. It will each 95 percent of 8963 (which is about 8765 ) when -yrr—lt##8be—r = 3765. Solving for Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.44 CONTENTS a PO) = oper Solving for K gives K =: 9211 = peceters nd f P(G0) = 450, then Geer = 450 ©. IE P(t) = ep ery = caer eset: and if P(60) = 480, then peer eaer = 480 Solving for K gives K 2 2664 £ Small differences in the 40-year projection result in huge differences in the estimated carrying capacity, D1.5.39 a. We have # = (0) = cv-dey _ pled) ‘Thus, the equation is separable. ‘Thus we can write S844 b, Wehave f (3 ‘Taking the exponential function of both sides gives « # de = J ($—d) de, s0by—alny = clnz—de+K, so de-+by = cz talny+K. = Cx°y® for an arbitrary constant C. D1.6 Chapter Review a a. False. It is a first-order, linear differential equation, but it isn't separable. b, Falso, It is a first-order, soparable differential equation, but it isn't linear. ©. True. Note that y! = 1 £7, so ty! = t—t-8, Thus, ty!-+y = t—t-8 + (t+ 6-1) = 2. Also, y(1) 4. True. ¢. False, In general, Euler's method gives approximate solutions. 2 v(t) = ~8y, so fat = f(a) at, ‘Thus, Inly| = —3 + K, and exponentiating both sides gives y= Ce 3 y(t) = —2y +6, so f SS dt = f Lat, and therefore =} In| ~2y+6|_=t+ K. It follows that |~2y +6] = Ae #50 y(t) = Ce +3. pe 4 pl(e) = 4p + 8 = Alp +2), so [2 de = fAde, and thus n|p + 2| = de + K. Thus |p +2| = Ae, so p+2= et, and thus p(x) = Ce — 2 5 yl(t) = 2ty, so f Mat = J 2tdt, and therefore, In|y| = t? + K. It follows that y= Ce 6 y'(t) = FB, s0 fy VyOat = feVPat. Integrating gives 2V7 = AT + K, s0 y= VE+C, s0 ult) = (vE+ CP. Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.D16. CHAPTER REVIEW 45 7 WO = Ly. Integrating both sides with respect to £ gives In y| = tan-1(t) + K, so y = Cet "0 8 2yy!(x) = sine, so y? = —cosx + C, and thus y= +V¥O— cose. 9 fA dt = f(2e+ tat, so tan'(y) = +-+C, and thus y= tan(t? +¢+C) = gai. Integrating both sides with respect to gives In|2| = bn |!2+1|-+K. Thus 2 = CYP>FT, 1 fy'(t)dt = f(2t + cost) dt, so y(t) =P +sint +C. Because y(0) = 1, we have 1=0+0+C, so C= 1 Thus, y(t) = 2+ sint +1. —3t-+ K, and thus y~3 = Ce“, Because J(-8) dt, so Iny — oH 3 12 y(t 3(y— 3), so f YQ ae (0) =4, we have C = 1, Thus y 13 (22 ar= side, 0 njQ- 8) = we have -8 = Ce, $0 C=. Thus, Q= Se $8. + K. Thus, Q—8 = Cet. Because Q(1) = 14 py/(2) = 2, 90 f yy!(2) de = f xd, so y?/2 = 27/2+C. Because y(2) = 4, we have 8 =2+C, 0 C Thus, y? = 2? +12, and y= vEP+TD 15 uM y(t) = 1-8, Thus fu-V8ut(t) dt = ft dt. We have $u2/* = have 6= $+C, 90 C= §. Thus, w/ = 243, and w= (2/8 +3). +. Because u(1) = 8, we 16 f(siny)y'(e)de = fAede, so —cosy = 2x7 + C. Because y(0) = x/2, we have 0 = 0+ C, oC Tinis cos y = ~222, anid y = cos-1(—22"). 17 f(es)s(that = f fly. Thus, 5? = In(t-$2) $C. Because s(-1) = 4, we have 16 = 0-4 C, 0 In(t +2) +16, and s = ine FF IO. 18 f sec? 46"(x) dx = f Ardy. Thus, tang follows that @ = tan~¥(2r? + 1). 1? + C. Because 6(0) = x/4, we have 1=0+C, s0 C= 1. It 19 SASSSNANSNNAASS SS c. The solutions are increasing when 0< A <2. 4. The solutions are decreasing when A <0 or when A > 2. ce. The equilibrium solutions are y= 0 and y= Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.46 CONTENTS 20 © Te appears that lim yoo y(t) = 1 4. fe appears that lim, jg ult) == e. The equilibrium solutions are y= 1 and y= —1 a. uy = Le uy = 1+ f(OL1)(A) = 141/20 = 1.05. Also, uz = 1.05 + f(1,1.05)(1) = 1.05 + 047619 = 1.09762. Thus, y(1) # 1.05 and y(.2) + 1.09762. Db. uy = 1. uy = 1+ F(0,1)(05) = 1+ 025 = 1.025. Also, up = 1.025 + f(.05,1.025)(.05) = 1.04939. aus = 1.04939 + f(-1,1.04939)(.05) = 1.07821. ug = 1.07821 + f(15,1.07821)(05) = 1.0961. ‘Thus lL) x 1.04039 and y(.2) = 1.09651 vt . For part a, we have b is more accurate. = 0.00198539. For part b, we have 12%5=YT3 — 9,999972108. Par 0.00198539. For part b, we b ae 0,000972108. Part Copyright ©2011 Pearson Education, Inc. Publishing as Addison-Wesley.D16. CHAPTER REVIEW 22 ‘The equilibrium solutions are y = 0 (unstable) and 23 SAQSNSANANNAANS 24 ‘The equilibrium solutions are y = —n/2 (stable) and y= 0 (unstable), and y = 7/2 (stable). SK eee eee LLL Copyright © 2011 Pearson Edueation, Ine. Publishing as Addison-Wesley.