We already know how to evaluate ∫032tdt. It's the area under y=2t from t=0 to t=3, a triangle with base 3 and height 6, so it's 9. Now what happens if we replace the upper bound 3 with a variable x?
Here, x is where the shaded region stops, and t runs across the region from 0 to x.
g(x)=∫0x2tdt=x2−02=x2
21⋅x⋅2x=x2
Find the derivative of ∫0x2tdt and ∫1x1+t3dt
Text solution
The derivative is the height at the edgeThe area up to x is x2, and its derivative 2x is the height of y=2t at t=x.
g′(x)=2x
Start at 1 instead of 0Removing the fixed triangle from 0 to 1 takes off a constant, so the derivative is still 2x, not f(x)−f(1)=2x−2.
∫1x2tdtdxd(x2−1)=x2−1=2x
No antiderivative neededWe can't integrate 1+t3 by hand, but the derivative is just the height at x.
dxd∫1x1+t3dt=1+x3
Why is the derivative the height at the edge? Moving x by dx adds a thin strip of height f(x), so the area grows at the rate f(x), whatever the curve. Where the curve is below the axis, f(x)dx is negative and g decreases.
g(x+dx)−g(x)≈f(x)dx
g′(x)=f(x)
This is the Fundamental Theorem of Calculus, and it's the exact reason why F(b)−F(a) gives the net area.
dxd∫axf(t)dt=f(x)
2
When the bounds depend on x
When the top bound is just x, the derivative of ∫axf(t)dt is f(x). But what if the top bound is x2, or x is in the bottom bound, or both bounds move?
Differentiate an integral with bounds that depend on x
Text solution
Plugging in x2 isn't enoughEvaluating first gives 4x3, not 2x2. The extra factor 2x is the derivative of the top bound.
∫0x22tdtdxdx4=(x2)2=x4=4x3=2(x2)⋅2x
Plug in the bound, times its derivativeReplace t with x2 and multiply by 2x.
dxd∫1x21+t3dt=1+(x2)3⋅2x=2x1+x6
x in the bottom boundSwapping the bounds changes the sign, so the derivative gets a minus sign.
∫x31+t3dt=−∫3x1+t3dt
dxd∫x31+t3dt=−1+x3
Both bounds: split at a constantSplitting at 1 leaves one moving bound in each integral.
∫xx21+t3dt=∫1x21+t3dt−∫1x1+t3dt
dxd∫xx21+t3dt=2x1+x6−1+x3
That extra factor is the chain rule. With w=x2 as the inside function, the derivative with respect to w is f(w), and we multiply by dxdw.
dxdy=dwdy⋅dxdw
w=x2
dxdw=2x
dxd∫0wf(t)dt=f(w)⋅dxdw=f(x2)⋅2x=2(x2)⋅2x=4x3
When both bounds move, the region gains a strip at the top edge x2 and loses one at the bottom edge x. That's why the top bound's term is added and the bottom bound's term is subtracted.
3
Finding values from a graph
Now suppose we only know f from its graph. With g(x)=∫0xf(t)dt, the value g(x) is the running total of signed area from 0 to x, so we can read it from the shapes under the graph.
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
y
f
Find g(2),g(4), and g(−1) for g(x)=∫0xf(t)dt from the graph of f
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
y
f
Text solution
Start at g(0)=0The integral from 0 to 0 has width 0.
g(0)=∫00f(t)dt=0
Add the shapes to the rightOn [0,2] a trapezoid and a triangle add 3 and 2. On [2,4] both shapes are below the axis, so their areas 2 and 3 come off.
g(2)g(4)=3+2=5=5−(2+3)=0
Left of 0, swap the boundsThe integral runs backwards from 0 to −1, so it's the negative of the triangle's area 1.
g(−1)=∫0−1f(t)dt=−∫−10f(t)dt=−1
It's easy to see the triangle on [−1,0] above the axis and think g(−1) is positive 1. But moving x from −1 to 0,g goes up by 1 and lands on g(0)=0, so it had to start at −1.
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
y
f
−1
0
1
2
3
4
5
−1
0
1
2
3
4
x
y
(−1,−1)
(0,0)
(2,5)
(4,0)
Starting at 0 made g(0)=0. If we instead wanted an accumulation function that equals 5 at x=2, we would write 5+∫2xf(t)dt. At x=2 the integral is 0, and the derivative is still f(x).
5+∫22f(t)dt=5+0=5
What happens when the upper bound is left of the lower bound?
The running total of signed area
4
Finding where g increases and its extrema
Because g′(x)=f(x), the graph of f is the graph of g′. So the height of f at any x gives the slope of g, not the value of g. Let's use that to find where g increases and where its extremes are on [−1,4].
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
g′
f
−1
0
1
2
3
4
5
−1
0
1
2
3
4
x
y
g
Find where g is increasing, its relative maximum, and its absolute extrema on [−1,4]
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
y
f
−1
0
1
2
3
4
5
−1
0
1
2
3
4
x
y
Text solution
Increasing where f is positiveg increases where g′=f is above the axis, and decreases where it's below.
f(x)>0on(−1,2)
f(x)<0on(2,4)
The relative maximum is where f changes signAt x=1, where f is tallest, g′(1)=4 only means g climbs fastest there. At x=2,f changes from positive to negative, so g peaks there.
f:+→−atx=2
g(2)=5
Compare the critical point and the endpoints
g(−1)=−1
g(2)=5
g(4)=0
max=g(2)=5
min=g(−1)=−1
5
Concavity and inflection points from the graph of f
The height of f tells us where g is increasing and decreasing. To see where g is concave up or concave down, we take one more derivative of g′(x)=f(x).
g′(x)g′′(x)=f(x)=f′(x)
The height of f gives g′, and the slope of f gives g′′.
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
g′
f
f′=2
f′=−4
f′=2
−1
0
1
2
3
4
5
−1
0
1
2
3
4
x
y
g
concave down
concaveup
concaveup
Find where g is concave up and concave down, and its inflection points
−4
−3
−2
−1
0
1
2
3
4
−1
0
1
2
3
4
t
g′
f
−1
0
1
2
3
4
5
−1
0
1
2
3
4
x
y
g
Text solution
Concavity from the slope of fWhere f is increasing, g′′ is positive and g is concave up. Where it's decreasing, g is concave down.
f′>0on(−1,1),(3,4)
f′<0on(1,3)
Inflection points where f turnsConcavity changes where f′ changes sign, at x=1 and x=3. At x=2,f(2)=0 gives the relative maximum, but f is decreasing on all of (1,3), so the concavity doesn't change there.
g′′(2)=f′(2)=3−1−4−4=−4
The height of f and the slope of f can have completely different signs.
(1,2)
g′=f>0
g′′=f′<0
(3,4)
g′=f<0
g′′=f′>0
On (3,4),g is decreasing and concave up at the same time. A positive f only means g is increasing, not concave up.
Does g have its highest value where f has its highest value?
The height of f is the slope of g
Summary
The Fundamental Theorem of Calculus
In general
dxd∫axf(t)dt=f(x)
Example
dxd∫1x1+t3dt=1+x3
When the lower bound is a constant and the upper bound is plain x, replace t with x in f(t). You don't need to find an antiderivative. The constant lower bound doesn't add a term to the derivative, so don't subtract f(a).
When bounds depend on x
Upper bound
dxd∫0x2f(t)dt=f(x2)⋅2x
Lower bound
dxd∫x3f(t)dt=−f(x)
Both bounds
∫xx2f(t)dt=∫1x2f(t)dt−∫1xf(t)dt
When a bound is a function of x, plug that bound into f(t) and multiply by the derivative of that bound. If x is in the lower bound, swapping the bounds gives a minus sign. When both bounds depend on x, split the integral at a constant to get the term from the upper bound minus the term from the lower bound.
Finding values of an accumulation function
Starting value
g(x)=∫axf(t)dt,g(a)=0
Integrating backwards
g(−1)=∫0−1f(t)dt=−∫−10f(t)dt
An accumulation function gives the signed area from the lower bound a to x, so g(a)=0. Area above the axis adds to the total, and area below subtracts. For x<a, reversing the bounds adds a minus sign, so area above the axis subtracts. Adding a constant, as in C+∫axf(t)dt, makes the value at x=a equal to C.
First derivative and extrema of g
g′(x)=f(x)
The height of f gives the slope of g. Where f is positive, g is increasing, and where f is negative, g is decreasing. A relative maximum occurs where f changes sign from positive to negative, not where f reaches its highest point. A relative minimum occurs where f changes sign from negative to positive. To find the absolute maximum and minimum of g on a closed interval, compare the values of g at the endpoints and at the points where f changes sign.
Concavity and inflection points of g
g′′(x)=f′(x)
The slope of f gives the concavity of g. Where f is increasing, its slope is positive, so g is concave up. Where f is decreasing, its slope is negative, so g is concave down. Inflection points of g occur where f switches between increasing and decreasing, not where f(x)=0. Always decide concavity from whether f is increasing or decreasing, not from the concavity of f.
6
Practice
Enter exact values unless a problem asks for an estimate.
1
Find dxd∫1xcos(t2)dt.
The upper bound is x, so what do you get when you evaluate cos(t2) at t=x?
If you got cos(x2)−cos1 The lower bound is the constant 1. Its derivative is 0, so it doesn't add a second term.
The upper bound is x and the lower bound is a constant, so we evaluate the function inside the integral at x to get the derivative.
dxd∫1xcos(t2)dt=cos(x2)
2
Let F(x)=∫0x3e−t2dt. Find F′(x).
The upper bound is x3, not just x. What does the chain rule say to multiply by?
If you got e−x6 Don't forget the derivative of the upper bound. Since the bound is x3, multiply by 3x2.
If you got 3x2e−x2 Replace t with the whole upper bound x3, so t2 becomes (x3)2.
The upper bound is x3, so let w=x3 and find its derivative.
w=x3,dxdw=3x2
Plug x3 in for t and multiply by dxdw.
F′(x)=e−(x3)2⋅3x2=3x2e−x6
3
Find dxd∫sinx31+t41dt.
The variable bound sinx is on the bottom. What sign do you get when you reverse the bounds?
If you got 1+sin4xcosx When you reverse the bounds to put sinx on top, the integral picks up a negative sign.
The variable bound is on the bottom, so reverse the bounds and put a negative sign in front.
∫sinx31+t41dt=−∫3sinx1+t41dt
The negative sign stays out front. Plug sinx in for t and multiply by the derivative of sinx.
Split the integral into two pieces at 0. Which piece has x on the bottom?
If you got cos(9x2)−cos(x2) The upper bound of the first piece is 3x, so remember to multiply by the derivative of 3x.
If you got 3cos(3x2)−cos(x2) Squaring the upper bound 3x gives (3x)2=9x2, not 3x2.
If you got 3cos(9x2)+cos(x2) Check the sign of the second term. The bound x is on the bottom of the original integral, so its piece is subtracted.
Both bounds depend on x, so split the integral at 0 and reverse the bounds of the piece with x on the bottom.
H(x)=∫03xcos(t2)dt−∫0xcos(t2)dt
Take the derivative of each piece, using the chain rule on the first piece with upper bound 3x.
H′(x)=cos((3x)2)⋅3−cos(x2)=3cos(9x2)−cos(x2)
5
Let g(x)=∫0x1+t44−t2dt. On what intervals is g increasing, and on what intervals is it decreasing? At which values of x does g have a relative maximum or a relative minimum?
Find g′(x). Since the denominator 1+x4 is always positive, where is the numerator positive?
If you got a relative maximum at x=0g′(0)=4, which is positive, so g is increasing at x=0. A relative extremum happens where g′ changes sign, not where g′ is largest.
By the Fundamental Theorem of Calculus, g′(x) is the function inside the integral evaluated at x.
g′(x)=1+x44−x2
The denominator 1+x4 is always positive, so the sign of g′(x) matches the sign of 4−x2.
1+x4>0,4−x2>0on(−2,2),4−x2<0on(−∞,−2),(2,∞)
At x=−2,g′ changes from negative to positive, so g has a relative minimum. At x=2,g′ changes from positive to negative, giving a relative maximum.
x=−2:g′:−→+,x=2:g′:+→−
6
The graph of f on [−4,4] is shown. It starts with a semicircle of radius 2 below the t-axis from t=−4 to t=0, followed by a line segment from (0,0) to (2,2) and a line segment from (2,2) to (4,0). Let g(x)=∫0xf(t)dt. (a) Find g(−4) and g(4). (b) Find g′(3) and g′′(3). (c) Find the absolute maximum and absolute minimum values of g on [−4,4]. Justify your answers. (d) At which values of x does g have an inflection point?
−2
−1
0
1
2
−4
−3
−2
−1
0
1
2
3
4
t
y
f
f is negative on (−4,0). As x goes from −4 to 0, does g(x) increase or decrease?
If you got g(−4)=−2π The integral from 0 to −4 is the negative of the integral from −4 to 0. Since the region is below the axis, the two negative signs cancel to give a positive value.
Compare the values of g at the critical point x=0 and the two endpoints. Is 2π greater than 4?
If you got an absolute maximum of 4 at x=4 Check the value at the left endpoint. g(−4)=2π, which is greater than 4.
(d)
Inflection points of g happen where g′′=f′ changes sign. Where does f change from decreasing to increasing, or from increasing to decreasing?
If you got an inflection point at x=0f(0)=0, but f increases on both sides of 0, so f′ doesn't change sign there.
Values of gThe region from −4 to 0 is a semicircle of radius 2 below the axis, and the region from 0 to 4 is a triangle with base 4 and height 2 above the axis.
∫−40f(t)dt=−21π⋅22=−2π
g(−4)=∫0−4f(t)dt=−∫−40f(t)dt=−(−2π)=2π
g(4)=21⋅4⋅2=4
Derivatives at x=3By the Fundamental Theorem of Calculus, g′(3)=f(3), and g′′(3)=f′(3) is the slope of the line segment containing x=3.
g′(3)=f(3)=1,g′′(3)=f′(3)=4−20−2=−1
Absolute extremaBecause g′=f,g decreases on [−4,0] and increases on [0,4], so the minimum is at x=0 and the maximum is at an endpoint.
f<0on(−4,0),f>0on(0,4)
g(−4)=2π,g(0)=0,g(4)=4
2π>4:max=g(−4)=2π,min=g(0)=0
Inflection pointsInflection points occur where g′′=f′ changes sign, which is where f changes between increasing and decreasing.
f′<0on(−4,−2),f′>0on(−2,2),f′<0on(2,4)
x=−2,x=2
7
Let g(x)=5+∫2xt3+1dt. Use the tangent line to the graph of g at x=2 to estimate g(2.1).
What is the integral from 2 to 2?
If you got 0.3 Check g(2). The constant 5 out front means g(2)=5+∫22t3+1dt=5, not 0.
At x=2, the integral has matching bounds, so it equals 0.
g(2)=5+∫22t3+1dt=5+0=5
Find g′(x) with the Fundamental Theorem of Calculus, then evaluate it at x=2 to get the slope of the tangent line.
g′(x)=x3+1,g′(2)=9=3
Write the linear approximation using the value g(2)=5 and the slope g′(2)=3.
g(2.1)≈g(2)+g′(2)⋅0.1=5+3⋅0.1=5.3
8
Find h→0limh1∫11+ht2+3dt.
If g(x)=∫1xt2+3dt, what is g(1)? Rewrite the fraction as a difference quotient for g.
If you got 0 At h=0, the numerator is ∫11t2+3dt=0 and the denominator is 0, which gives 00, not 0.
Let g(x)=∫1xt2+3dt. Then g(1)=0, so the expression is the difference quotient for g at 1.
g(x)=∫1xt2+3dt,g(1)=0
h1∫11+ht2+3dt=hg(1+h)−g(1)
The limit of the difference quotient is g′(1), and by the Fundamental Theorem of Calculus, g′(1) is the function inside the integral evaluated at 1.
h→0limhg(1+h)−g(1)=g′(1)=12+3=2
9
Find x→0limx3∫0xsin(t2)dt. (Challenge)
As x approaches 0, what do the numerator and denominator approach?
What is the derivative of ∫0xsin(t2)dt with respect to x?
If you got 1 The derivative of the denominator x3 is 3x2, so keep the factor of 31.If you got 0 The upper bound is x, not x2, so the derivative of the numerator is sin(x2) with no extra factor of 2x.
Plugging in x=0 gives 00, so we can use L'Hôpital's rule.
x=0:∫00sin(t2)dt=0,03=0
Differentiate the numerator and the denominator with respect to x.