Scribd
Upload
388 views

Strauss - Chapter - 4

Uploaded by

R.V. Raghavan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
388 views

Strauss - Chapter - 4

Uploaded by

R.V. Raghavan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 37
46, Use the MVT to evaluate cose +1 Lim, 47. Let f(x) = 1+ ~. Ifa and b are constants such that a <0 and b > 0, show that there is no number w between a and b for which JO) ~ fa) = Fwy a) 48. Show that for any x > 4, there is.a number w between 4 and x such that 1 rod” 2Yo Use this fact to show that if x > 4, then ve 15, Him: Let f(x) = VIF Use the MVT to show that 0ex <2 jan x. Note that Jia) = f0) =0 Show that there is no number w between 0 and =r for which {f'(an) = 0, Why does this fact not contradict the MVT? Use Rolle’s theorem or the MVT to show that there is no num. bera for which the equation ea3rta=0 thas rivo distinct solutions inthe interval [—1, 1}. Ifa > Disa constant, show that the equation i fpar—1=0 thas exsetly one real solution, Hint: Let f(x) =. +ax = 1 ‘and use the intermediate value theorem © show that there is ‘a least one root. ‘Then assume there are two roots, and use Rolle’s thevicmi (@ obtain a contradiction For constants « and b, a > 0, and na positive integ Rolle’s theorem or the MVT to show that the polynor use bart pe cam have at most one real root Show that if f(x) = 0 forall x, then f is a linear function, (That is, (4) = Ax +B for constants A #0 and B.) Show thatif /"(s) = Ax-+2 forconstants A % Oand B, then F(x) is a quadratic function, (That is, f(x) = ax? + bx +e for constants a,b, and c, where a #0.) 4.3 Using Derivatives to Sketch the Graph of a Function IN THIS SECTION increasing and decreasing functions, the frst-derivatve tes, concovily and inflection points, the second-derivative test, curve sketching using the first and second derivatives Our next goa! is to see how information about the derivative f” and the second deriva- tive f” can be used to determine the shape of the graph of f. We begin by showing how the sign of is related to whether the graph of f is rising or falling 202 Ghopter 4 Aditonl Aplications ofthe Derivative INCREASING AND DECRE: (G FUNCTIONS Suppose an ecologist has modeled the size of a population of a certain species as a function f of time ¢ (months). IF it turns out that the population is increasing until the end of the first year and is decreasing thereafter, it is reasonable to expect the population to be maximized at time ¢ = 12 months and for the population curve to have a high point at ¢ = 12, as shown in Figure 4.15. If the graph of a function f, such as this population curve, is rising throughout the interval 0 < x < 12, we say that f is Strictly increasing on that interval. Similarly, the graph of the function in Figure 4.13 strictly decreasing on the interval 12 <1 < 20. These terms may be defined more formally as follows: Strictly Increasing and Strictly Decreasing Functions oD [ / ef) (a) {t + at 1-4 Figure 4.15. population curve | i IN ee CaaS la | icant a ees ial “al ‘Function sly increasing 1. Function sry desreasing Figute 4.16 Increasing and decreasing functions Note thot we donot use the word “strictly” every time we tolk cbout.o A function fis said to be (strictly) monotonic on an interval / if it is either siriely monotonic fonction, or a strciy ly increasing on all of F or strictly decreasing on all of f, The monotonic increasing function, or o striely de- is closely related to the sign of the derivative "(x). In particular, if the graph of a creasing function, function has tangent lines with positive slope on /, the graph will be inclined upward and f will be incteasing on I (see Figure 4.17). Since the slope of the tangent at each WARNS 2 note on twoge: We say thot functions ore inereasing ond that rophs are rising, For exomple, if f(x ?, we soy that the function f° icicle bs ees eae groph is rising for x > 0. Figue 4.17 ‘The graph is rising where f’ > Oand falling where f° < 0. ‘Notice that the small lags indicate the slope at various points on the graph. 4.3 Using Desinovesto Ske eGrph ofa Fucion 203, point on the graph is measured by the derivative /’, itis reasonable to expect f to be increasing on intervals where f’ > 0. Similarly, it is reasonable to expect f 10 be decreasing on an interval when f’ < 0. These observations are established formally in ‘Theorem 4.7, THEOREM 4.7 Monotone function theorem Let f be differentiable on the open interval (a, b). If f/x) > Oon (a, b), then f is strictly increasing on (a, 6), If f/(x) < Oon (a,b), then f is strictly decreasing on (a, b). Proof We will prove that f is strictly increasing on (a, 6) if f"(x) > 0 throughout the interval. The strictly decreasing case is similar and is left as an exercise for the reader. Suppose ”(x) > 0 throughout the interval (a, 6), and let x; and x be two num- bers chosen arbitrarily from this interval, with x1 < x2. The MV'T tells us that Fla) = fed) _ moe Fle) or fa) — fe) = ser — x) for some number ¢ between x) and x2. Because both f'(c) > O and x2 — xy > 0, it follows that f’(c)(x2 — x.) > 0, and therefore fla)—flny>0 or flr) > fon) That is, if x; and x are any two numbers in (a,b) such that x4) < x2, then F(x) < fas), which means that fis strictly increasing on (a,b). a To determine where a function f is increasing or decreasing, we begin by finding the critical numbers (where the derivative is zero or docs not exist), These numbers divide the x-axis into intervals, and we test the sign of "(x) in each of these intervals. If f’Cx) > Oinan interval, then f is increasing in that same interval, and if f"(x) <0 in an interval, then f is decreasing in that same interval To indicate where a given function f is increasing and where itis decreasing, we | mark the critical values on a number line and use an up arrow (}) to indicate an interval where fis increasing and a down arrow ({) to indicate an interval where f is decreasing, Sometimes, when the full graph of f is displayed, we indicate the sign of {f" on an interval bounded by critical numbers with a string of + signs if f is increasing ‘on the interval ora string of — signs if f is decreasing there. This notation is illustrated in the following example. EXAMPLE 1 jing intervals of increase and decrease Determine where the function defined by f(x) = x3 —3x?—9-+1 is strictly increasing and where itis strictly decreasing, Solution First, we find the derivative: 3x? — 6y —9 = 3(x +) —3) 1 numbers. f(a) exists for all « and f"() — 0 at x= —land ‘These critical numbers divide the x-axis into three parts, as shown in Figure 4.180, and we select any arbitrary number from each of these intervals. For example, we select —2, 0, and 4, evaluate the derivative at these numbers, and mark ceach interval as increasing (+) or decreasing (1), according to whether the derivative positive or negative, respectively. This is shown in Figure 4.18b. The function f is sing for x < —1 and for x > 3; f is decreasing for —1 0, what can be said about the graph of f? >. When the graph of / is falling, what can be said about the graph of {”? cc. Where do the critical numbers of ff appear on the graph of f? Solution ‘The graphs of f and ” are shown in Figure 4.19. a, When f"(x) > 0, the graph of f is rising. This occurs when x < —I and when x23, b. When the graph of / is falling (for —1 < x <3), we have f"(x) < 0, so the graph of is below the x-axis. . The critical numbers of f are where f"(x) = 0; that is, at x = —1 and x = 3, 50 they are the x-intercepts of the graph of f. . ‘THE FIRST-DERIVATIVE TEST Every relative extremum is a critical point, However, as we saw in Section 4.1, not every critical point of « continuous function is necessarily a relative extremum, If the derivative is positive to the immediate left of a critical number and negative to its im- mediate right, the graph changes from increasing to decreasing and the critical point must be a relative maximum, as shown in Figure 4.20s. If the derivative is negative to the immediate left of a critical number and positive to its immediate right, the graph changes fiom decreasing (o increasing and the critical point is a relative minimum (Figure 4.20b). However, if the sign of the derivative is the same on both immediate sides of the critical number, then it is neither a relative maximum nor a relative min- imum (Figure 4.20c). ‘These observations are summarized in a procedure called the Jfirst-derivative test for relative extrema, OTT) Wena Range, OFS) mae) (eee i ji ifep, (fea reh) pee Th 0] \ 3 ese PN A relative maximam 4.4 relative minimum ce Noeateeaum Figue 4.20 Theve patterns of behavior near a critical number 4.3 Using Derive fo Shot the Goph of aFuncion 205 Step 1. Find all eritical numbers of a continuous function f. That is. find all mumbers ¢ such that f() is defined and either fc) = 0 or J"(e) does, not exist. Step 2. Classify each ical point (c, f(¢)) as follows: ‘a, The point (c, f(0)) is a relative maximum if /\ F's) > 0 (graph rising) for all x in an open interval (a..} to the left of ¢, and #'(8) < 0 (graph falling) for all x in an open interval (c, b) to the right of ¢. b. The point (c, f(e)) is a relative mi \ i #2) < 0 (graph falling) for all x in an open interval (a,¢) to, the left of c, and f'(&) > 0 (graph rising) for all x in an open anterval (c, 6) 10 im if } i the right of c. ox |e. The point (c, fc) is not an extremum if the derivative /"(x) has \ \ the same sign for all x in open intervals (a,c) and (c, b) on each side of e ‘Suppose we apply this first-derivative test to the polynomial f(x) = x3 — 3x? — ‘9x 1 In Example 2 we found that this function has the critical numbers —1 and 3 and that f is increasing when x < —I and x > 3 and decreasing when —I 0, which means that the graph of f is concave up where the second derivative f" satisfies f” > 0. Simi larly, the graph is concave down where f” < 0, We use this observation to characterize concavit 4.3 Using Deitives Skok th Graph of Funcion 207 | Derivative Characterization of Concavity ‘The graph of a function f is concave up on any open interval 7 where J's) > 0, and it is concave down where /"(x) <0. When discussing the concavity ofa function f. we will display a diagram in which a “cup” symbol (U) above an interval indicates that f is concave up on the interval and a “cap” symbol (9) indicates that f is concave down there. This convention is illustrated in the following example. EXAMPLE 4 Concavity for @ polynomial function Find where the graph of f(x) = x° + 3x + 1 is concave up and where it is concave down. Sotution We find that f(x) = 312 +3 and /%(x) = Ga. Therefore, /"(x) < Oif.x < O and J" (x) > Oifx > 0, so the graph of {fis concave down for x < 0 and concave up for Figue 4.24 Intervals of concavity x > O, as indicated in Figure 4.24. Returning to Figure 4.22c, the graph is concave ‘down to the left of x = 0 and concave up to the right. Hence, this is the graph of f(x). . In Figure 4.23, notice that the graph changes from concave up to concave down at the point C. It will be convenient to have a special name for such transition points. Inflection Point maa An inflection point oan on 1, meaning fe) must Reel at econ poet et x Returning to Example 4, notice that the graph of f(x) = x° + 3x + [has exactly: ‘one inflection point, at (0, 1), where the concavity changes from down to up. Various kinds of graphical behavior are illustrated in Figure 4.25, Note that the graph is rising on the interval (a, ci], falling on [¢1,c2}. rising on [¢2, es}, falling on [es, cal, rising on (4, c5}, ain falling on (cs, 6]. The concavity is up on [. pals and down otherwise. In this figure, and elsewhere, when the graph of / is displayed, we IL indicate the sign of {f” on an interval by a string of + signs if f” > 0 on the interval, and a string of ~ signs iff” < 0 there “The graph has relative maxima at ¢), ¢3, and ¢s, and relative minima at c: and ¢4. ‘There are horizontal tangents (f"(x) = 0) at all of these points except at ci and 3, ‘where there are sharp points, called corners (fe) and J"(c3) do not exist). There is a horizontal tangent at py; that is, /"(pi) = 0, but no relative extremum appears there. Instead, we have points of inflection at p, and p, because the concavity changes direction at each of these points. In general, the concavity of the graph of f will change only at points where “f'x) = 0 or f(x) does not exist—that is, at critical numbers of the derivative f(x). We will call the number c a second-order critical number if f"(c) = 0 or {"() does not exist, and in this context an “ordinary” critical number (where f'(c) = 0 or if'(c) does not exist) will be referred to as first-order critical number. If we do not Specify otherwise, a critical number is always a first-order critical number. Inflection points correspond to second-order critical numbers and must actually be on the graph 208 Chopter 4 Atona Applications of the Derivative i t 4 { I Figure 4.26 ‘The graph of f(x) = x* ‘has no inflection poimt at (0, 0) even though "(0) = 0, WARNING. 9 continuous hietion f need not have on inflection point ot every number ¢ where f"(c) = fix) 4 we have 0, but he graph of Fis always concave up (see Figure 4.26). For Rejatwe aim tne point tang | Tatangelo _|_ [Lm poly anna Figure 4.25 A graph of a function showing critical points and inflection points of f. Specifically, a number ¢ such that f"(c) is not defined and the concavity of f changes at ¢ will correspond to an inflection point if and only if /(c) is defined. Here is an example of one way inflection points may appear in applications. EXAMPLE 5 Peak worker efficioncy An efficiency study of the moming shift at a factory indicates that the number of units produced by an average worker: hours afier 8:00 A.M. may be modeled by the formula Q() = —F +91? + 12r. At what time in the moming is the worker performing most efficiently? Solution We assume that the morning shift runs from 8:00 A.M. until noon and that worker efficiency is maximized when the rate of production R(t) = Q(t) = ~ 3? + 181 + 12 is as large as possible for 0 = t < 4. The derivative of R is, RQ) =O") = -6r +18 which is zero when t = 3; this is the critical number. Using the optimization criterion of Section 4.1, we know that the extrema of R(t) on the closed interval (0, 4] must ‘occur at either the interior critical number 3 or at one (or both) of the endpoints (which are 0 and 4), We find that RQ) = 12 RG) =39 RG) =36 so the rate of production X(t) is greatest and the worker is performing most efficiently when ¢ = 3; that is, at 11:00 A.M. ‘The graphs of the production function Q and its derivative, the rate-of- production function &, are shown in Figure 4.27. Notice that the production curve is steepest and the rate of production is greatest when 1 = 3. 4.3 Wing Dero tS the Graph af aFunton — 209 ‘A production curve bs, Rate of production Figue 4.27 Graph of a production curve showing the point of diminishing returns Ml In Example 5, note how the rate of production, as measured by the slope of the graph of the average worker's output, increases from 0 to the inflection point J and then decreases from / to E, as shown in Figure 4.27. Because the point / marks the point where the rate of production “peaks out,” it is natural to refer to 7 as a point of diminishing returns. 11 is also an inflection point on the graph of Q. Knowing that this point occurs at 11:00 A.M., the manager of the factory might be able to increase the overall output of the labor force by seheduling a break near this time, since the production rate changes from increasing to decreasing at this time. THE SECOND-DERIVATIVE TEST Itis often possible to classify acritical point P(c, f(c)) on the graph of f by examining the sign of f”(c). Specifically, suppose f'(c) = Oand f"(c) > 0. Then there is a horizontal tangent line at P and the graph of f is concave up in the neighborhood of . This means that the graph of f is cupped upward from the horizontal tangent at P, ‘and it is reasonable to expect P to be a relative minimum, as shown in Figure 4:28a, Similarly, we expect P to be a relative maximum if f"(c) = 0 and f"(c) < 0, because the graph is cupped down beneath the critical point P, as shown in Figure 4.28b, fep=0 T [reso + tit Tt 2, Relative minimum by. Relative maxima Flo) =Oand (> Plo) =Oand fe) <0 Figure 4.28 Second derivative test for relative extrema ‘These obsetvations lead to the second-derivative test for relative extrema. eee The Second-Derivative Test for Relative Extrema Let f be a function such that f"(c) = 0 and the second derivative exists on an ‘open interval containing ¢. If f"c) > 0, there isa relative minimum at « If f"(c) <0, there isa relative maximum at x = If f%(c) =0, then the second-derivative test fails (either a maximum, or a minimum, or neither may occur). —— 210 Ghopter 4 Aan! Apis of he Deiative Direction: Ns ISe(x-l)(441), Negative 0 Negative Figure 4.30 The graph of EXAMPLE 6 Using the second-derivative test Use the second-derivative test to determine whether each critical number of the fun tion f(x) = 325 — 5x3 +2 comesponds to a relative maximum, a relative minimum, or neither. Solution Once again, we begin by finding the first and second derivatives: FG) = 1x4 = 15x? = 157% — NO +1) F(x) = 60x3 — 30x = 30x(2x? — 1), Solving f/(x) = 0, we find that the critical numbers are x = 0,.x = 1, and.x = =I We plot these points on a number line, as shown in Figure 4.29, To apply the second- derivative test, we evaluate f"(x) at each eritical number. 70) fa) test fails at x = 0, 30 > 0; positive, so the test tells us that there is a relative minimum at S(H1) = —30 < 0; news atx ve, so the test tells us that there Is a relative maximum, . ‘When the second-derivative test fails, as at x = 0 in Example 6, the critical point can often be classified using the first-derivative test. For instance, the first derivative in Example 6 is Fe) = 15° = DOF D We see f(x) = atx = —1, 0, and 1, We can plot these points on 2 number line, shown in Figure 4.29, and then evaluate f"(x) at test numbers just to the left and just to the right of each eritical number. ring | faltng falling rising ib a eas ee Figure 4.29 Sign graph for the derivative of f(x) = 3x° — 5x +2 We show a number line like this as part of the graph of many of our examples (as we did in Figure 4.25), but most often we show the derivative to the right and left of critical number. For Example 6, the derivative is negative both to the immediate left ‘and right of 0, which we illustrate as shown in Figure 4.30. Neither kind of extremum occurs at x = 0 ({ | pattern). EXAMPLE 7 Finding inflection points Find the inflection points for the function f(x) = 3x° — 5x? + 2, given in Example 6. Solution ‘We begin with the second derivative (from Example 6): 30x(2x? = 1) To find the inflection poims, we look at the second-order critical numbers, namely, We can show these on a sign graph of the second derivative, as in Figure 4.31. Notice that the inflection points ‘occur where the sign of the second derivative changes. The graph of f is shown in Figure 4.32, along with the sign graphs for both the first and second derivatives. Spend some time studying this figure to sce how the signs of the first derivative indicate where the graph is rising and where itis falling, as well 4.3 Wing Deivovesia Skethbe Graph ofan 217 ee pe eee Figue 4.31 Sign graph for the second derivative of f(x) = 31° — Sx?-42 how the signs of the second derivative show the concavity. For completeness, the figure also shows the sign graph for the function itself. The graph is above the x-axis where f is positive, below the x-axis where / is negative, and crosses the x-axis where sO) =0. Figure 4.32 Graph of f(x) = 3x? —5x° +2 . Example 7 demonstrates both the strength and the weakness of the second-derivative test. In particular, when itis relatively easy to find the second derivative (as with a poly- nomial) and ifthe zeros of this function are easy to find, then the second-derivative test provides a quick means for classifying the critical points. However, if it is difficult to compute f%(c) or if "(c) = 0, it may be easier, or even necessary, to apply the first-derivative test CURVE SKETCHING USING THE FIRST AND SECOND DERIVATIVES Now, we are ready to develop a procedure for curve sketching, The key ideas appear in Example 8, EXAMPLE 8 Sketching the graph of a polynomial function Determine where the function f(x) = a4 —4x° + 10 is increasing, where itis de- creasing, where its graph is concave up, and where its graph is concave down. Find the relative extrema and inflection points, and sketch the graph of f Solution ‘The first derivative, Sx) = 48 = 12s? = 47x — 3), is zero when x = 0 and when x = 3, Because 4x? > 0 for x #0, we have f'(x) < 0 for x ~ 3 (except for ¥ = 0) nnd f"() > O for x > 3. The pattem showing where fF is increasing and where itis decreasing is displayed in Figure 4.330, Next, to determine the concavity of the graph we compute Sf") = 12x? — 24x = 12x — 2) ify 2, then f"(x) > O and the graph is concave up. It is concave down when 0 0, so there 0, as indicated in Figure 4.35 We find that the second derivative of f i a relative maximum at Ne Lo which is zero when x = £1. We find that f(1) = f(—1) ~ 0.24, and that the concavity of the graph of / is as indicated in Figure 4.35b, Finally, we draw a smooth curve through the known points, as shown in Figure 4.36. ‘The graph of /f rises to the high point at approximately (0, 0.4) and then falls indefinitely, approaching the x-axis arbitrarily closely because e~"*/? approaches 0 as jx] increases without bound. Note that the graph has no x-intercepts, because & always positive, (04) flrs pointarst- EXAMPLE 10. Sketching the graph of a trigonometric function Sketch the graph of T(x) = sin.x + cos.x on (0, 277] Solution ‘You probably graphed this function in trigonometry by adding ordinates, However, with this example we wish to illustrate the power of calculus to draw the graph. Thus, ‘we begin by finding the first and second derivatives. T(x) = ens x —sinx T(x) = — sin — cos We find the critical numbers (both 7” and 7” are defined for all values of x): =i and.x = % and x= in v2 thus, T/(x) = O when cos.x 17"(x) = Owhen cos.x = —sin.x; thus, x = The intervals of inerease and decrease as well as the concavity pattem are shown in ye 4.37. 214 Ghopter 4 Aditenl Applian of he Deratire Shapes} “| C7 | OX Direction: i a ate 1 Intervals of crease bb. Concavity pattern ‘nd decrease Figure 4.37 Preliminary work for sketching T(x) = sin.x + eos Find the critical points, the points of inflection, and the endpoints . Relative maximum: 7 (3) = V3; the exitical point is , V3). Relative minimum: (3) = —V2; the critical point is (2, - V3), Inflection: TE 0; the inflection point is (3, 0) T (22) = 0; the inflection point is (2% , 0). Endpoints: 7(0) = 1 and Tx) = L Finally, pass a smooth curve through these key points as indicated in the preliminary sketch in Figure 4.384 to obtain the completed graph shown in Figure 4.38b. Pins Beemer ! Figure 4.38 Graph of T(x) = sinx ++ cos.x on 10. 21 . ! 4.3 PROBLEM SET WHAT DOES THIS SAY? What isthe first derivative test? 2. Exploration Problem What i the relationship between the graph ‘ofa function and the graph of its derivative? 3. WHAT DOES THIS SAY? What is the secondlerivative test? 4. WHAT DOES THIS SAY? What is the relationship between concavity, points of inflection, and the second derivative? 5. Exploration Problem ‘The cartoon on page 209 exclaims, “Our prices are rising slower than any place in town.” Restate using caleulus o In Problems 6-7, identify which curve represents a function f and which curve represents its derivative J Draw a curve that represents the derivative ofthe function defined by the curves shown int Problems 8-11 oh Por the functions in Problems 12-19, (a. Find all critical numbers, 'b, Find where the function is increasing and decreasing. €, Find the critical points and identify each as a relative maxi- ‘muna, velative miniman, or neither d. Find the second-order critical numbers and tell where the _graph is concave np and where its concave down. fe. Sketch the graph. 1 fiy=t 942 1B fl =a +L 14. fle) =a + Set — 5500? ~ 2.0001? + 60,000x. 1S. f(x) = x9 $3542 — 125x — 9,375 L 16, fan 11, f= 355 18 f(x) = In 9. fu) =U 41-5) 1 Problems 20-35, determine the intervals of increase and de- crease and concavity for the given function, and then use those intervals to help you sketch its graph. 2. fc) = —12)' =20°= 12° 2, fos) = 142+ 18/r 22. fu) = ut — 20? — 12a? +180 23, gu) = 0° + 60! — 24? +26 4. fix) = VE TT 25. 90) Deas ja 26. Fy =P 4307 B. f= 1int Bw. sy =re™ ete 3H. fx) = ans? 9 — 2e0s6 for 0.< 0 < 20 0 +0820 for0 <0 <7 f(x) = 2x — sin"! for—} sx <1 Sf) In Problems 36-39, use the first-derivative test 10 classify each of the given cvtical nuunbers as a relative miniuum, a relative maxi ‘muna, or neither 36. f@) = (0 —304 I) tes x= al x? sin on [- 4.3. Using Dnvtives Sethe GrophofoFoncion 215, 39. fx) = YOR atx = {In Problems 40-43, use the second derivative test 10 classify each of the given critical munbers as a relarive ttinimaum, a relative maxinuwn, or neither 40. Fo) de) $30 —12v4 Wate = 1x 2 x45 vee (P= 3r teats 1xed in + Jose atx (4 Eaporaton Problem Skech the graph of «function withthe following properties: 41. fx) ay =.= P(e) >0 when x <1 Flay >0 fay <0 Fray <0 srey>o when x > 3 when when alere3 x<2 when x >2 45, Exploration Problem Sketch the graph of a function with the following properties: x<2 andwhen 20 when when when when x<2 andwhen 47 46. Exploration Problem Sketch the graph of « funetion with the following properties: Sey >0 fa <0 fey>0 fe) >0 ‘when when when when rel xol vel xo ‘What can you say about the derivative of f when x = 1? 47. Sketch the graph of a function with the following properties: “There are relative extrema at (—1, 7) and (3.2). There is an inflection point at (1,4). The graph is concave down only. ‘when x < 1. The x-intercept is (—4, 0) and the y-intercept is 0.5). 48, Sketch a graph of function f that satisies the following con- ditions: @ f@) > Owhen.x <—S and when. > 1. Gi) f() Owhen =I 3. Gi) (C1) =Oand £°G) = 0. 216 Chapter 4 dean ppication af the Deiat 50, In physies, the Formula (2) wher 10) ‘and Jy is a constant, is used to model light intensity in the study of Fraunhofer diffraction a, Show that £(0) = lo. b, Sketch the graph for [3.3m]. What are the critical pois on this interval? 51, Ata temperature of T (in degrces Celsius), the speed of sound in iris modeled by the formula 103 1) v=up/t+ 35 where vp is the speed at 0°C. Sketch the graph of v for 7 > 0, nd use calculus to check for ritical points. Modeling Problems: In Problems 52-53, set up an appropriate ‘model 1o answer the given question. Be sure to state your assump- '52, Atmoon on a certain day, Frank sets out to assemble five stereo sets, His rate of assembly increases steadily throughout the af- ternoon until 4:00 vat, at which time he has completed three sets. After that, he assembles sets at a slower and slower rate until he finally completes the fifth set at 8:00 Pt, Sketch a rough graph of a function that represents the number of sets Frank has completed after hours of work. 53, An industrial psychologist conducts two efficiency studies at the Chiloo appliance factory. The first study indicates that the average worker who arrives on the job at 8:00 A.M. will have assembled =P 46413 blenders in r hours (without a break), for 0 < 1 <4, The second study suggests that after a 15-minute coffee break, the average worker can assemble benders in hours after the break for 0 < ¢ <4, Note: The 1Ssminute break is not part ofthe work ime. A. Verity tha if the cafe break cute at 10:00 4.24, the a= erage worker will assemble 42 blenders before the break and 49! blenders forthe two hours after the break. i, Suppose the coffee break i scheduled to begin x hours af= ter 8:00 A.M. Find an expression for the total number of blenders V(x) assembled by the average worker during the morning shift (8.8.6. to 12:15 P.M.) ‘At what time should the coffee break be scheduled so that, the average worker will produce the maximum number of blenders during the morning shift? How is this optimum, time related to the point of diminishing retums? 54 Rosoorch indicator that the power P required by a bird to maintain flights given by the formula where v fs the relative speed of the bird, w is its weight, p is the demity of aix, and Sand A are constants associated @ 56. HISTORICAL QUEST One of the with the bird's size and shape.* What speed will minimize the power? You may assume that v, wa, S, and A are all positive. 55. The deflection o hardwood beam of length is given by De) where x is the distance from one end of the beam, What value of x yields the maximum deflection? most fomous women in the history of mathematics is Mario Goétana Agnesi {pronounced din y8'22). She wos bern in Milon, the fist of 21 children. Her first publication was at age 9, when she wrote a Latin discourse defend- ing higher education for women. Her most important work was a now-dlasse calculus textbook published in 1748. ‘Maria Agnes! is primarily remembered fr o curve defined by the equation MARIA AGNES! 1718-1799. eta ‘The curve was named versiera {rom the loan ver f turn) by Agnes, but John Colson, an Englishman who translated her ‘work, confused the word versiera with the word owersiera, “which means “wife of the devil in talon; the curve has ever Since been called the "witch of Agnesi” This was particularly unfortunate because Colton wanted Agnes's work lo sorve.os ‘a model for budding young mathematicians, especially young ‘women. Graph this curve and find the critical numbers, extrema, and points of inflection, 57. Journel Problem: Mathematics Magazine’ tary proof that a apositive constant Give an elemen- is positive and inereasing $538. An important formula in physical chemistry is vam der Was equation, which says that (r+ Rr vi where P, V, and 7 are the pressure, volume, ancl temperature, respectively, of a gas, and a, b, m, and R are positive con- stants. The eritical temperature Te of the gas is the highest temperature at which the gaseous and liquid phases can exist as separate states, ‘a, When T = Te, the pressure P can be expressed as a func- tion #(V) of ¥ alone, Show how this ean de done, and then find P’(V) and P"(V), ¥G.5. Fennycuick, “The Mechanies of Bird Migration." bi I (1969), pp. 525-556, * Volume 55 (1982), . 300, “Elementary proo!”in the question means that yousshould use only techniques from beginning calculus. For our purposes, You simply need to give a veasonable argument to justiy the conchison, b The critical volume Ve is the volume that satisfies P'(Vc) = Oand P"(Ve) = 0. Find Vc Te, the point where P"(V) = 0, using the Vo from part b to write it in terms of a,b, and R. Finally, find the critical pressure Pe = P(Vc) in terms of a, bm, and &. d. Sketch P asa function of V '59, Prove or disprove that if the graphs of the functions f and g are both concave up on an interval, then the graph of their sum f+ isalso concave up on that interval 460, Use caleulus to prove that for constants a, b, and ¢, the vertex (relative exitemum) ofthe quadratic function yeartbrte (#0) occurs atx = —b/ (2a). 61, Find constants A, B, and € that guarantee that the funetion: Fi) =AP FBC HC 2. 63. 444 Carve Skthing with sympltes Lis ivan finty 207 will have a relative extremum at (2, 11) and an inflection point at (1, 5), Sketch the graph of f Find constants a, b, and c that guarantee thatthe graph of fae tar?toxte will have a relative maximum at (3, 18) and a relative mini mum at (1,14). Exploretion Problem Consider the graph ot y =P box? hex} d for constants b, ¢, and d. What happens to the graph as 6 changes? Find constants A, 8, C, and D that guarantee that the graph of fO) = 34+ AP + BP +Cx4D ‘will have horizontal tangents at (2, ~3) and (0,7). There isa third point that has a horizontal tangent. Find this point. Then, forall three points, determine whether each corresponds to @ Felative maximum, a relative minimum, or nother 4.4 Curve Sketching with Asymptotes: Limits Involving Infinity IN THIS SECTION limits to infinity, infinite limits, graphs with asymptotes, vertical tangents and cusps, general graphing strategy LIMITS TO INFINITY In applications, we are often concerned with “long run” behavior of a function, To indicate such behavior, we write lim so =L to indicate that f(x) approaches the number L as x increases without bound. Similarly, ‘we write lim, £0) = to indicate that f(x) approaches the number M as x decreases without bound. Here are the formal definitions of these limits to infinity. Limits to Infinity WARNING, Een though the symbols co and +00 mean te some thing, for Hes thew Let wer vse 4-00 te help distinguish between +00 ond —co. ‘This definition can be illustrated graphically, as shown in Figure 4.39. 218 chopter 4 AddiinalAppeations of the Dette at fim ji) = Figure 4.39 Graphical representation of limits to infinity With this formal definition, we ean show that all the rules for limits established in ‘Chapter 2 also apply to lim, F(x) and , lim, F(x). ‘The following theorem will allow us to evaluate THEOREM 4.8 Special limits to infinity TA eae eee ‘number, then Furthermore, i is suc tat 2” is defined for x < 0, then lim 3 = Proof We begin by proving that lim x > Nwehave isa Figure 4.40 Graph of ae etes) Y> 55428 =7 44 Cae Sting thymine iiniy, 219 This means that | — 0} < ¢, 0 that, ftom the definition of mit wehave, Tim x : Now let be a rational number, say» = p/q. Then The proof for the analogous limit as x > —o follows similarly When evaluating a limit of the form PAs) pix) Por tim F* ate d(x) sett d(x) where p(x) and d(x) are polynomials, itis often useful to divide both p(x) and d(x) by the highest power of x that occurs in either. The limit can then be found by applying ‘Theorem 4.8. This process is illustrated by the following examples. EXAMPLE 1 Evaluating a limit to infinity ‘ 30 = 5x49 Eviloate Ue S$ 22 7 Solution We may assume that x # 0, because we are interested only in very large values of x. Dividing both the numerator and denominator of the given expressions by x°, the highest power of x appearing in the fraction, we find, 1 Be =Sx4+9 _ 3x - 5x49 SO FRP 7 Sx 42x77 Thus, “540-0 5 "The graph of the given rational function is shown in Figure 4.40, Notice how the curve seems to approach y = $ as x —> -+oo and as x > —00, . Example 2 illustrates how Theorem 4.8 can be used along with the other limit properties to evaluate limits to infinity, EXAMPLE 2 Evaluating li Evaluate 220° choptor 4 dtl ppeaons of th Devatine Solution Notice that forx #0, Also, according to Theorem 4.8, we know that 3 2 lim |==0 and fim = =0 We now find the limits using the quotient rule, the power rule, and Theorem 4.8: et (=a = wrt 2) Le i ( lim (1— Similarly, EXAMPLE 3. Evaluating a limit to negative infinity fm 29% +57 +30 Fratuate Tm = 1,000 Solution Dividing the numerator and the denominator by the highest power, x5, we find that i 95x 457e430_ 95x34 57430 GS eee pee oe — = i _0+0+0 1-0 Fig 4.41 Graph of . jac Sea The graph of the rational function is shown in Figure 4.41, . EXAMPLE 4 Evaluating a limit to infinity involving e Find lime cosx. Sotution We cannot use the product rule for mits since im eos.x does not exist (it diverges by oscillation —do you see why?). Note, however, that the magnitude of & cosx = must become sinaller and smaller as x > +00 Since the numerator cos x is bounded ie } between —I and 1, while the denominator e" grows relentlessly lager with.x. Thus by | the squeeze rule, slime" cose = 0 osx is shown in Figure 4.42, . Figure 4.42 Graph of y=e-*cosx ‘The graph of y = Infinite Limits 44 Carve Stthing with symptotsLinisvching inty, 221 NABNING, Tey drawing the graph of y =e cns.x on your coleuoter, or other graphing software, and you wil see that what you get does not propely illustrate the oxilalory behavior of ths graph. This demonstrates the importance of nat relying on technalagy for graphing. In foct, the accompanying graph (without scale) shows more cleerly the behavior ofthe graph of the function y= e-* COS x fr large, positive values of x INFINITE LIMITS {In mathematies, the symbol 00 is not « number, but is used to describe either the process of unrestricted growth or the result of such growth. Thus, a limit statement such as lim f(x) = +00 ‘means that the funetion f increases without bound as x approaches c from either side, while Jim g(x) = =o ‘means that g deereases without bound as x approaches ¢. Such limits may be defined formally as follows. >} What This Says Remember, oo is not a number, so an infinite limit does not exist in the sense that limits were defined in Chapter 2. However, there are several ways for a limit not to exist (for example, lim cos.x fails to exist by oscillation), so saying that lim f(x) = +08 or lim f(x) = —99 conveys more information than simply observing that the limit does not exist. EXAMPLE 5 inite limits eee Solution Notice that L increases without bound as x approaches 2 from the right and = decreases without bound as x approaches 2 from the left. That is, and tim werd li rath x We also have lim @x — 5) = 1. and it follows that 3x-5 and tim 222 Vertical and Horizontal Asymptotes ‘Choper 4 Adina Aplin fhe Deraine GRAPHS WITH ASYMPTOTES Figure 4.43 shows a graph that approaches the horizontal line y = 2 as x —> -+oo and 1as.x > 00, and the vertical ine x = 3 as x approaches 3 from either side. Figuce 4.43 A typical graph with asymptotes jing a rational function with asymptotes Br—5 EXAMPLE 6 Graph Sketch the graph of (0) Solution Vertical Asymptotes First, make sure the rational function is written in simp] 3x-5 (reduced) form. Because vertical asymptotes for f(x) = occur at values for whieh tim _f() oF lim, (x) Is infte, we look for ‘aes that cause the nominator Be zor (and ths numerator n6t ta be 2aroys that i, we solve aCe) where d(x) is the denominator of f(x), and then evaluate tim f(x) and tim, f(x) ascertain the behavior ofthe function atx =. For this example, x = 2 isa value ‘causes division by zer9, so we find and (We found these limits in Example 5.) This means that x = 2 is a vertical asymptote _ and that the graph is moving downward asx —> 2 from the left and upward as x —» 2 recorded on the preliminary graph shown in Figure 4.dda by a dashed vertical line with upward (t) and downward ({) arrows. Horizontal Asymptotes To find the horizontal asymptotes we compute Bros wile 52 = lim 4A Cove Satin wih Asympes Lins lrehig kindy 223 and . (The stops here are the same as for. > +00.) ‘This means that y = 3 is a horizontal asymptote. This information is recorded on the preliminary graph shown in Figure 4.44a by a dashed horizontal line with outbound >) A Concafe —Hrakiaen = | Be Ssympiote Faling ae iad ear |. Concave owe bs, Completed sketch 3x-5 Figure 444 Graph of f(x) = 2 ‘The preliminary sketch gives us some valuable information about the graph, but it does not present the entire picture. Next, we use calculus to find where the function is increasing and decreasing (first derivative) and where it is concave up and concave down (second derivative): swe a Neither derivative is ever zero, and both are undefined at x = 2. Checking the signs Of the first and second derivatives, we find that the graph is always falling and that it concave up for x > 2 and concave down for.x < 2. However, it does not have a point of inflection (the funetion is not defined at x = 2). This information is added to the preliminary sketch shown in Figure 4.44a. The completed graph is shown in Figure 4.44b, which also shows the x- and y-intercepts at (5, 0) and (0, 3). . EXAMPLE 7 Sketching a curve with asymptotes aox-2 Discuss and sketch the graph of f(x) = Solution We find that 6r+5 . 8 gp POS Gas Solving /"(x) = 0, we'see thatthe critical numbers are x = 1 and.x = 5, Testing on cach side of the critical numbers and 3, where f(x) is not defined, we obtain the Jnizrvale of ipesoon gel erento ancl camcaiy elowa in igor 4 ‘Note that there is a relative maximum atx and a relative minimum at 5. ‘The concavity changes (from | to ¢) at x = 3, but this does not correspond to an inflection point since (3) is not defined, We look for vertical asymptotes where f(x) (in reduced form) is not defined; in this case, at x = 3. Testing with limits, we find that fa) 224 Chopter 4 Akon Applicaton of the Derive nema iraiita Tosline 1 Negnive 3 Negative 5 Posiive a. Intervals of increase and decrease bi Intervals of concavity Figure 4.45 Intervals of increase and decrease and concavity for f(x) = To check for horizontal asymptotes, We compute and ‘Since neither limit of f(x) at infinity is finite, there are no horizontal asymptotes. ‘We plot some points: the relative maximum (1, 1); the relative minimum (3, 9); the x-intercepts (2, 0), (—1, 0); and the y-intercept (0, 3). We also show the vertical asymptote x = 3, and, using the intervals of increase and decrease and concavity indicated in Figure 4.45, we obtain the graph shown in Figure 4.46, . VERTICAL TANGENTS AND CUSPS | ‘Suppose the function f is continuous at the point P where x = cand that f"(x) ‘becomes infinite as x approaches c. Then the graph of has a vertical rangent at P if the graph turns smoothly through P and a cusp at P if the graph changes direction sibruptly there. These graphical features can be defined in terms of limits, as follows: Vertical Tangents and Cusps These possibilities are shown in Figure 4.47. | sea ‘Vertical angemt he Vertical tangent 6 Cusp limp @)=t— limps tim) = + Eee terol Figue 4.7 Vertical wingents and cusps 44 Cave Sktding wih Asyplts Lins nang lfnty 225 EXAMPLE 8 Identifying vertical tangents and cusps In cach of the following cases, determine whether the graph of the given function has a vertical tangent or a cusp: a FO) 28Qr 45) b. g(x) =x"8(x 44) Solution aw iting f(x) = 285 4.50223, we find that fi) = B77 44 0. Since a+ = so f"(x) becomes infinite only at lim FG) = lim Pa a and im, fe) = Tim Bx! + 1) = Jim, £6) = Jim, x'8¢e + 1) = 400 it follows that there is a cusp on the graph of f at (0,0). The graph is shown in Tiguie 4.48a, b. The derivative e'@) = $x 41) ). We find that lim gf) = tim 4-27 + 1) = +00 Bao becomes infinite only when x and dim, a!) = fim 3x77 + 1) = +00 Since the limit approaches +00 as x approaches zero from both the left and the right, we conclude that a vertical tangent occurs at the origin (0,0). The graph is shown in Figure 4.48b, Figur 448 A graph with a cusp and another with a vertical tangent A GENERAL GRAPHING STRATEGY Itis worthwhile 10 combine the techniques of curve sketching from calculus with those techniques studied in precalculus courses. You may be familiar with extent (linding the domain and the range of the function) and symmetry (with respect to the x-axis, y-axis, or origin). These features ure reviewed in Chapter 8 of the Student Mathematics Handbook. We now have all the tools we need to deseribe a general procedure for curve and this procedure is summarized in Table 4.1. 226 Chapter 4 AditiorlAppcatons ofthe Deraive TABLE4.1 Graphing Strategy for a Function Defined by y = f(x) ET, ‘Step Procedure Fe) = BEES amu ai wf) = 2 beng ar find deat Comte is end ssn ees eho mo sey Bee. acai Sack acl fd asad endo c mbe peer rer Uae enbga rusian oC): qadlenas #°(8) > O, curve rising (indicate these regions by 1) F(a) <0, curve falling (indicate these regions by 1) Apply the second-derivative test. Use the second-derivative test find the relative maxima or minima; Substitute the first-order critical numbers, 1, with the following test: $"(8) > 0, relative minimum £74) <0, relative maximum (3) =0, test fails; use fist-erivaive test Determine concavity and Find the points of inflection, These are located where the concavity changes from up to down or of inflection. down to up, and are found by checking the intervals on either side of the second-order critical numbers. Use the first-derivative test if the second. derivative test fails or is too complicated. 1. Lat cy bea first-order critical number of a continuous function J. 2 ae (Ci, (CED) is arelative asians it r\ {F'G) = O forall xin an open interval (a, 4) tothe left ofc: {F'@) < O forall x in an open interval (cy, b) tothe right of . Mowimemy p(t, (ey) is avelative minimum if \/ FC) < 0 forall x in an open imerval (a, tothe left of F/G) > 0 forall x in an open interval (c;,b) to the right of cy. Minimum & (cr, (ci) is not an extremum if f(x) has the same sign in open imervals (a, c4) and ci.) on both sides of ey, Find asymprotes, vertical 1. vertical asymprores—Vertical asyinpwotes, i any exist, will occur at values x —¢ for which tangents, and cusps. isnot defined, Use the limits lim f(x) and lim_F() to determine the behavior ofthe graph near.x = ¢. Show this behavior with arows (71). 2. Horizontal asymptotes—Compute lim f(s) and associated horizontal asymptote. 3. Vertical iangents and cnsps—Find the vertical tangents and cusps: the graph has a vertical tangent at P if fim f"(x) and fim, f(x) are either both +00 of both 60, It has a cusp at P if im fC a nf). IFeithers finite, plot the lim, #"(x) are both infinite with opposite signs. Plot points. 1. If f(@) isa relative maximum, relative minimum, or inflection point, plot (c, f(c)). Show the aximum points by using 3 “cap (6) and relative minigun points by using a “cup” ©, 2 s-intercepes: Set y-intercepts: Set ‘one y-intercept Oandif x =a is a solution, plot (a. Oand if y = isa solution, plot (0, a function will have, at most, the curve. Draw a curve using the above infor 4.4 PROBLEM SET — © 1. wuarvoEs mis sav? Outine a method forcurve sketch ing 2, WHAT DOES THIS SAY? What are critical numbers? Dis- cuss the importance of critical numbers in curve sketching. 3. Exploration Problem Discuss the importance of concavity and points of ificetion in eatve sketching. 4, Exploration Problem Discuss the importance of asymptote ‘curve sketching. Evaluate the limits in Problems 5-24, 915 +5007 +800 1,000 tn (ssn 21 tin, BYE 22 ig BSE ta 23, tim. sin aa Find all verical and horizontal asympiotes of the graph of each “function given in Problems 25-44. Find where each graph is rising ‘and where itis faling, determine concavity, and locate all critical ‘poims and pois of inflection. Finally, skerch the graph. Be sure to show any special features, such as cusps or vertical tangents. aes as 28. fix) = 26. gx) = 2. fo) =442 m8. y=1- 75 2 2x? Sx +7 vw. j= St 30. 0) = 1 3h. gtx) = Sa ee 33. gi) = (PO +7 & shart je 38. S00 = 7-9)? 6. glx) = x(x? = 12) 3. fu sx 4) 38. fa = 0-7) 39 f(x) = tan! x? 40. f(x) = In = 27) 41. (0) = sind — cos? for 0 <0 <2 42. f(s) =x —sin2e for e~? 47, Sketch 4 graph of a function f with all the following prop- erties: The graph has y = 1 and x = 3 as asymplotes, f is increasing for x < 3and 3 < x < 5 and is decreasing elsewhere; the graph is concave up for x < 3 and for x > 7 and concave down for 3 3; (Gi) The graph has only one erties! point (1, ~1), no infec tion points; 6) im, 96a) =2: 0. 8) = im ge) 49. Exploration Problem Frank Komerkutter has put off doing his ‘math homework until the last minute, and he is now trying to evaluate At first he is stumped, but suddenly he has an idea: Because L lin, — = +00 00 and it must surely be true thatthe limit in question has the value +400 ~ (+00) = 0. Having thus “solved” his problem, he cel- ebrates by taking a nap. Is he right, and, if not, what is wrong ‘with his argument? in physics itis shown that if one travels around the curve at the ideal speed, no frictional force is required 10 prevent slipping. Thi greatly te- duces wear on tes and conbutes to safety. 228 Chapter 4 Aionl Aplin of he Deinotve 50. nan experiment, a biologist introduces a toxin into a bacterial colony and then measures the effect on the population ofthe colony, Suppose that a ime ¢ (in minutes) the population is Pe =ste "4D thousand. AL what time will the population be the largest? What happens to the population in the long run (as 1 > +20)? FFind where the graph of P has an inflection point, and inter- pret this point in terms of the population. Sketch the graph of P. 51, Let P(x) = ay" Hag x"! +0 ayx-hay be a polynomial with g, # Oand let C= tim P(x) and M= tim PO). Fill in the missing entries inthe following table: Signotan tt ig Mt + even 00 + oid be = een a - odd 00 ‘52. a. Show that, in general, the graph of the function atthe te re parte FO) ‘will have y = © as a horizontal asymptote and that when ‘br + as the graph will eros this asymptote atthe point where aed 1 Sete the graph of each ofthe following futons: wraae-5 Bxt-x-7 ssp “cypress . Find constants « and b that guarantee that the graph of the function defined by arts fay = S48 will have a vertical asymptote at x = $ and a horizontal asymptote at y = —3. 54, Journal Problem: Purabola.” m Draw a careful sketch of the nlcating clearly any vertical or horizon- {al asymptotes, tuning points, or points of inflection $5. HISTORICAL QUEST The possibility of division by zero. is faci thot causes special concern to mathematicians. One of the first recorded observations of division by zero comes from the twelti-century Hindu mathematician Shaskaracharya {also known as Bhaskara), who made the following observe tion: “The fraction whose denominator is zero is termed an infinite quanti.” Bhaskaracharya then went onto give a very beautiful conception of infinity tha involved his view of God “Volume 29, Isue 1 (1984) QCounterexample Problems in Problems 56 and creation. Bhaskorachorya gave « solution for the so: called Pell equation Le py? which is related fo the problem of cuting a given sphere so the volumes ofthe two porls have a specified ratio. Solve this equation for p = 2/3 and write this Pell equation acy — f(s)! Find lim. f(s) x, both nor 59, either show that the statement is generally rrue or fd « counterexample. 56, If f is concave up and g is concave down on an interval 1, then f¢ is neither concave up nor concave down on 1 57. IF f and g are concave up on the interval J, then so is f +8. 58. If f(x) > O and g(x) > 0 for all x on J and if f and g are concave up on /, then fis also concave up on 1 59. If f(x) < Oand fx) > 0 forall x on J, then the function = fis concave up on 1 160, Exploration Problem State what you think should be the formal definition of each of the following limit statements: b. lim f(x) 661, Consider the rational function gx ayant pee tax tay LOO a gE EB EB a I'm > m and by 0, show that the x ‘horizontal asymptote of the graph of f. bh It m= a, show tat he line y = a/bq isthe only hor polation dotnet If < mist posible forthe graph to havea horizontal a eee oe 2, Prove the fllowen is fim g(x) = A (4 > 0), then is the only. rule, If lim f() = +00 and Tim f@}g@] = +00 Hint: Notice that because limg(x) =A, the function (3) is near A when x is near c. Therefore, because “lim, £2) = +00 te product (1) g(3) i large if is near 6, Forralize these observations for the proof, 63. Prove that if tim_ f(a) and lim, g(x) both exist, so do lim (78) + GO] and Aa Ye aes eae in Te ey sti 6)= 1 « $for> and [g(2) = M| < $ for > No, then whenever x > for some number 1, IF) + gO (L +a] <€ ‘You should also show that AV relates to Ny and 13, “From "Mathematics in Indi inthe Middle Ages” by Chandra B, Sharma, Morhemarieal Speetram, Nolume 18(1). pp. 6-8. 1982. 4.5 VHépital’s Rule IN THIS SECTION MARNING, First check the (rmoke sure itis of the form 0/0 or 00/00}, and then take /"/g’, not sy. 45 [Hipitl’stule 229 «rule to evaluate indeterminate forms, indeterminate forms 0/0 and 0/20, other in- determinate forms, special limits involving e* and In. A RULI 10 EVALUATE INDETERMINATE FORMS In curve sketching, optimization, and other applic ate amit ofthe form tim £2. a(x ons, itis ofien necessary to evah- where lim (x) and fim g(x) ate ether both 0 or both 06. Such limits are called 0/0 indeterminate forms and 00/00 indeterminate forms, re- speetively, because their value cannot be determined without further analysis, There is 4 rule to evaluate such indeterminate forms, known as l'Fdpital’s rule, which relates rues) the evaluation toa computation of fim <7. the mit ofthe ratio ofthe derivatives of oe a {f and g. Here is a precise statement of the rule THEOREM 4.9 |'Hépital's Rule Let f and g be differentiable functions with g'(x) # 0 on an open interval containing (except possibly atc itself). Suppose lim a produces an indeterminate form $ or & and that The theorem also applies to one-sided limits and to limits at infinity (where x > oo and x + -00). Proof The proof of an important special cas cam obtain a sense of why it is true in the followin, are differentiable functions such that f(a) = g(a’ formula for a differentiable function F given in Appendix B. However, we uument. Suppose f(x) and g(x) 0. Then, using the linearization P(x) © Fla) + F(a)(x — 4) See Section 3.8. fe) _ J+ Sle =a) _ OF FWVe-a) _ fo) BQ) gla) +a'(ay(a—a) ~ O+gahix—a) ga) ° are) a 1 Hopital’s rule is named after Guillaume Frangois Antoine de P Hopital (1661~ 1704), the author ofthe first textbook on differential calculus, published in 1696. How- ver, the result that bears I’ Hépital’s name was actually due to Johann Bernoulli (see the Historical Quest in Problem 58) 230 Chapter 4 inal Application of the Devine INDETERMINATE FORMS 0/0 and 00/00 We now consider a variety of problems involving indeterminate forms. We begin with a limit first computed in Chapter 2, but instead of using a geometric argument together with the squeeze rule of limits, we use 'Hépital’s ru EXAMPLE 1 Using 'Hépital's rule to compute a fa Evaluate tim “"*, ‘ es Solution ‘Note that this is an indeterminate form because sin.x and.x both approach 0 as x —> 0. This means that 'Hopital’s rule applies: sinx lim = EXAMPLE 2 I'Hépital's rule with a 0/0 form pes. =8" Evaluate lim Solution For this example, f(x 8, and the form is 0/0. Hépitals rule Simplify L 2212 es Limit of a quotient . EXAMPLE 3 Limit is not an indeterminate form cos Evaluate tim rid setx Solution ‘You must always remember to check that you have an indeterminate form before ap- plying "H6pital’s rule. The limit is Tease | SRC o0s3) 6 lig OO ‘ x0 seer lim see x 1 2 WARNING, 1 you blindly apply 'Hepital’s rule in Example 3, you obtain the WRONG answer This is NOT correct. a See x tan x = lim ra I seex 1 This onswer is blatantly WRONG, as you can see by looking ot the Technology Note. “thor he lip yee from te sight 45° Wipil’sfule 231 a graphing calculator to help find many indeterminate-form limits. For = instance, the limit in Example 3 can easily be checked by looking at the graph. If you \V have ng calculator, you can see that as x — O from either the ieft or the right, the limit looks the same, namely, 0, YiBCi-cos Xcos Xmin= 76, 283185, Kmax=6. 285188: Rs 576736, min: omax: Ysel EXAMPLE 4 I'Hépital's rule applied more than once asin Evaluate lim Solution ‘This is a 0/0 indeterminate form, and we find that ylookng at she arp you can renorce pears nile EXAMPLE 5 I'Hépital's rule with an oo/co form 22 3x41 B im valuate Rat Sx 2 Solution Using the methods of Section 2.2, we eould compute this limit by multiplying by (/x)/(1/22). Instead, we note that this is of the form 90/00 and apply I’ Hopital’s rule 3 23x41 ae Arp apres ig sesh MM ey se ply pitt cule sea 2 3 Te may happen that even when 'Hépital’s tule applies (0 a limit, i is not the best way to proceed, as illustrated by the following example. EXAMPLE 6 Using I'Hépital’s rule with other limit properties (1 =cos.x) sine 3 eos 32 Copter 4 ‘itil glo fh Deave WARNING, 'Hépital's rule ise cppies only to the indeterminate form 0/0 and c0/o0. Solution This limit has the form 0/0, but direct application of I"Hépital’s rule leads to a real mess (try it). Instesd, we compute the given limit by using the product rule for limits first, followed by two simple applications of I"Hopital’s rule, Specifically, using the product rule for limits (assuming the limits exist), we have vim (L=ooswysinds _f,, L=cosx Jf singed, suo cose [SO a? rad x | Leo cosx Tin 28°) Pin 2©°°°] Pig = LR oe | LET 2 cosx. - :) ay) =2 . =(;)@w= EXAMPLE 7 Hypotheses of 'Hépital's rule are not satisfied Evaluate tim **88" sotto ¥ cos Solution This limit has the indeterminate form co/co. If you try to apply 'Hopital’s rule, you find 1+cosx im sotto Tsing ‘The limit on the right does not exist, because both sinx and cos. oscillate between =l and 1 asx —> +00. Recall that I’Hopital’s rule applies only if tim £ ae +00, This does not mean thatthe limit ofthe original expression does hot exist or that we cannot find it; it simply means that we cannot apply I H6pital’s rule. To find this it factor out an.x from the numerator and denominator and proceed as follows: Loris tim “sins 304M) La 40 A Neer res a OTHER INDETERMINATE FORMS. Other indeterminate forms, such as 1%, 0°, 06°, co — 00, and 0 00, can often be ma- nipulated algebraically into one of the standard forms 0/0 or 0/00 und then evaluated using I'Hépital’s rule, The following examples illustrate such procedures. We begin by deriving a formula given without proof in Section 2.4, EXAMPLE 8 Limit of the form 1~ 1 Show that fim, (+4) Solution Note that this limit is indeed of the indeterminate form 1°. Let tin (14), i Leck a te raph a x Fram ata hole andthe ng. 45° lMépil's Rule 233, Take the logarithm of both sides: Ink =f tn, (1 + ‘)] Inx fe continuous Property of logarithms (Hépitats rule simplify aa Thus, nL = Land L =e! =. . EXAMPLE 9 I'Hépital's rule with the form 0 - 00 ant evatite tin, (s—2)rans Solution Ths limita ts form 0 0 bens lim (x— 0 and lim tanx = +00 son V2 some 1 ; Write tan.x = —— to obtain cole x xo¥ 0 lim (x—%)tany = tim Form = De ee ee a = tim tHépitat’s rule sore Sex uae = lim (—sin* x) . roe EXAMPLE 10 Limit of the form 0 Find tim x". ra Solution ‘This is a 0” indeterminate form. From the graph shown in Figure 4.49, it looks as though the desired limit is 1. We ean verify this conjecture analytical 234 Copter 4 Aol Ape: he Dervae Figure 4.49 Graph of 1°" Asin Example 8, we begin by using properties of logarithms. L= lim xt" In = In lim, x** xs = Him tn.c* Inis continuous This 15 0+ 60 form, This is form, % ee THOpiears rule at = C86. X COL re sin? x im sas OSE (8) (<) M@=0 W ‘Thus, L =e? = EXAMPLE 11 Limit of the form oo” Find tim x", ste Solution ‘This is a limit of the indeterminate form oo. WL = him x!/*, then InL = In fim x'/* = lim nx" the limit of a log is the log of the limit. xan ie = jim tinx alee 0 5 | This is 2 form = IHépitals rule ‘Thus, we have InL = 0; therefore, L. EXAMPLE 12. Horizontal asymptotes with 'Hépital’s rule Find all horizontal asymptotes of the graph of y = xe Solution To test for horizontal asymptotes, we compute Figue 4.50 Graph of y = xe-* showing the x-axis a «horizontal asympiote ae) 45° (pits Rule 235, and for the second, =0 Thus, y = 0 (the x-axis) is a horizontal asymptote. The graph of y = xe-** is shown, in Figure 4.50, . EXAMPLE 13 'Hépital’s rule with the form co ~ co Balt tim ( a Solution 1 As it stands, this has the form oo —~oo, because ~ > -o0and —— + +o0asx > 0 from the right. However, using a little algebra, we find al 14 sinx = -90 lim ** =0 a ote gm & note lim =+00 lim x"e* =0 eb rato 236 Chopter 4 Aionl pian of the Daivatve Proof ‘These tan all be verified directly or by applying I'H6pital’s rule. For example, Inx lim = xoHoe ‘The other parts are left for you to verify. a > What This Says The limit statements 7 et Inx em ee portant. They tell us that, in the long run, any exponential e* dominates any power x", for & and n positive, which, in turn, dominates any logarithm, 4.5 PROBLEM SET © 1. Anincorect use of Hopital’ rule is ilasrated nthe follow ing timit computations. In each ease, explain what is rong and find the corect value ofthe imi lease = 14 cote 24) fy HOO a+ 50 26. lim_see3xcos9x im = a i tim, 3% 0 i 28. im Sin 2. kxplortin Problem Sometimes IHpital’s rule leads to faulty 29. tim x? Ine 30. lim (sinx)Inx ‘computations. For example, observe what happens when the an os rule i applied to 31. tim 202d By aim (eg) a mmol 35, ime! +)" 36, im (sin 2)! QO Find cach of the timity in Problems 37-51. a. 38. 3, fm, = 1] Hint: Ine* 40, tim x = Ine + €°*)] Hint: Ine* sinas tande al 1 tin 8 ao 0 12, tig SE 41. im, (4-1v2) 42. lim(e! ~1-3)" 13 tig — 14, tim $= 43, im, Inx)(cotx) 44, tin ce — ane —x "an aM AM 1 tin BSE singe 446. tim SOESIND Fras eT 17 tm SBSESINE 10 sinae ial 20, tim. sin ~ er ©. Bi im. [in(#)-2425] $82, Find A so that Ii In Problems 53-56, use I'Hopital’s rule to determine all horizon: tal asymprotes to the graph of the given function, You are NOT required to sketch the graph Ins’ Exploration Problem Write a paper on using technology to eval: sate limits. HISTORICAL QUEST The French mathematician Guil- Jaume de I'Hapital is best known today for the rule that bears his name, but the rule was discovered by I'Hopital’s teacher, Jehenn Bernoulli, Not only did IHépitel neglect to cite his sources in his book, but there is also evidence that he paid Bernoulli for his results and for keeping their arrangements for payment confidentcl In o letter dated Morch 17, 1694, he asked Bernoulli "to communicate to me your discoverias «with the request not fo mention them fo others". would not pleose me iF they were made public.” UHépital's ‘argument, which wos originally given without using functional 53. fx)= 54, fa) = 55. fo) = (nO 37. 58. notation, can easily be reproduced? flatds) _ flay+ f(ardx s(atdx) ~ gta)+ eta) a Flardx Flaydx Supply reasons for this argument, and give necessary eondi- Gions forthe functions f and g. HISTORICAL QUEST. The remorkable Bernoulli foily of Switzerland produced atleast eight noted mathematicians over three generations. Two brothers, Jacob (1654-1705) and Jo- hann (1667-1748), were bitter rivals. These brothers were extremely influential advocates of he newly born calculus. Jo: honn was the most prolific of the clan ond was responsible forthe discovery of Hépitals rule (soe Problem 58), Bemoul numbers, Bernoulli polynomials, the lerniscate of Bernoulli, the Bernoulli equation, the Bomoull theorem, and the Bernoul 59. SD. d, Serik. A Source Book tn Maremattes, 1200-1800, Cambridge, MA: Hacvard University Pres, 1969, pp. 313-316. 45° (Hopi fule 237 distribution. He did o great deol of work wih differential equa ions. Johann was jealous and cantankerous; he tossed a son, {Doni cul hs how forwrning en ward he had exeeed towin himself. « Write a report on the Bernoulli family Q Find te tis in Problems 60-62 using the following methods. graphically —b analytically ¢, numerically Compare, contrast, and reconcile the three methods. 60, lim x 66. or. 61. im") ae Ol elf Find all values of A and # so thot . Fora certain value of C. the limit lim (4 + 51° +3) — x is finite and nonzero. Find C and then compute the limit For which values of constants D and £ is it rue that lim(s sins + Dx* + £) = ~2? ‘A weight hanging by a spring is made to vibrate by applying ‘sinusoidal force, and the displacement at time 1 is given by (sina — sin) where C, a, and A are constants such that a . What hap- peens fo the displacement as > 7 You may assume that a iefived . For F and G positive constants, define fay =e + FP & Compote Ly = lim f(x) and Le = im. £6), a b, What is the largest value of F for which the equation Ly = GLy has a solution? What are Ly and Ly in this

You might also like