I dont Know

Latticeee

May 31, 2025

Abstract
This document provides a compilation of common integral rules, some advanced integration
techniques, and a selection of challenging or ”almost impossible” integrals, along with hints for their
[Link] try these before the exam.

Common Integral Rules Fundamental Theorem of Calculus

Rx
1. If F (x) = a f (t) dt, then F ′ (x) = f (x).
Basic Rules
Rb
•
R
k dx = kx + C 2. a f (x) dx = F (b) − F (a), where F ′ (x) = f (x).

xn+1
• xn dx =
R
n+1 +C (n ̸= −1)
Advanced Integration Tech-
• 1
R
dx = ln |x| + C
x niques
• ex dx = ex + C
R
Integration by Parts
ax
• ax dx =
R
ln a +C R R
Formula: u dv = uv − v du.
•
R
sin x dx = − cos x + C Choose u and dv, find du and v, then apply. Often
useful when integrating products of functions (e.g.,
•
R
cos x dx = sin x + C
xex , x sin x, ex cos x) or functions like ln x, arctan x.
• sec2 x dx = tan x + C
R

•
R
csc2 x dx = − cot x + C Trigonometric Integrals
sinm x cosn x dx
R
Integrals of the form or
•
R
sec x tan x dx = sec x + C
secm x tann x dx.
R

•
R
csc x cot x dx = − csc x + C
• If n (power of cos x) is odd, save one cos x factor
• and use cos2 x = 1 − sin2 x. Let u = sin x.
R
tan x dx = ln | sec x| + C = − ln | cos x| + C
• • If m (power of sin x) is odd, save one sin x factor
R
cot x dx = ln | sin x| + C
and use sin2 x = 1 − cos2 x. Let u = cos x.
•
R
sec x dx = ln | sec x + tan x| + C
• If both m and n are even, use half-angle identi-
•
R
csc x dx = ln | csc x − cot x| + C ties: sin2 x = 1−cos(2x)
2 , cos2 x = 1+cos(2x)
2 .
• For tanm x secn x dx:
R
Inverse Trigonometric Functions
– If n (power of sec x) is even, save sec2 x and
• √ 1 x
R

a2 −x2
dx = arcsin+C a use sec2 x = 1 + tan2 x. Let u = tan x.

• a2 +x
R1 1
2 dx = a arctan
x

– If m (power of tan x) is odd, save sec x tan x
a +C
  and use tan2 x = sec2 x − 1. Let u = sec x.
• x√x12 −a2 dx = a1 arcsec |x|
R
a +C
Trigonometric Substitution
√ √
2 2 2 2
Properties of Integrals √ for integrands involving a − x , a + x ,
Used
or x2 − a2 .
•
R R
kf (x) dx = k f (x) dx √
• For √a2 − x2 : let x = a sin θ, dx = a cos θ dθ.
•
R R R
[f (x) ± g(x)] dx = f (x) dx ± g(x) dx Then a2 − x2 = a cos θ.

1
 √
• For √ 1
a2 + x2 : let x = a tan θ, dx = a sec2 θ dθ. 8. sin4 x+cos
R
4 x dx
2 2
Then a + x = a sec θ. R∞ 2
√ 9. 0 e−x dx (Gaussian Integral)
• For x2 − a2 : let √x = a sec θ, dx = R∞
a sec θ tan θ dθ. Then x2 − a2 = a tan θ. 10. 0 sinx x dx (Dirichlet Integral)
Remember to draw a reference triangle to convert
back to x. ”Almost Impossible” Integrals (Often
requiring special functions, complex
Integration by Partial Fractions analysis, or advanced techniques)
P (x) R1 ln(1+x)
Used for integrating rational functions Q(x) . 1. dx (Sophomore’s Dream related)
0 1+x2

1. If degree of P (x) ≥ degree of Q(x), perform long 2. R π/2 ln(sin x) dx

0
division first.
R ∞ ln x
2. Factor the denominator Q(x) completely. 3. 0 (1+x2 )2 dx
R∞
3. For each factor in Q(x), write a term in the par- 4. Frullani’s Integral: 0 f (ax)−fx
(bx)
dx (General
tial fraction decomposition: R ∞ e−ax −e −bx
form) For example: 0 x dx
• Linear factor (ax + b): A
ax+b
R 1 arctan x
5. 0 x√1−x2 dx
• Repeated linear factor (ax + b)k : A1
ax+b + R∞ xs
A2 Ak 6. 0 (x+a)(x+b) dx (Mellin Transform related)
(ax+b)2 + · · · + (ax+b)k
R∞ cos(ax)−cos(bx)
• Irreducible quadratic factor (ax2 + bx + c): 7. 0 x2 dx
Ax+B
ax2 +bx+c R∞ 1

dx
8. The ”Integral from Hell”: 0 1+x2 − cos x x
• Repeated irreducible quadratic factor
A1 x+B1 Ak x+Bk (Regularization needed)
(ax2 +bx+c)k : ax 2 +bx+c +· · ·+ (ax2 +bx+c)k
R∞ a c
9. 0 xs−1 e−x cos(b)−x sin(d) dx (Very general, of-
4. Solve for the constants A, B, Ai , Bi .
ten no closed form)
5. Integrate the resulting simpler fractions.
10. ln1x dx (Li(x), Logarithmic integral function,
R

non-elementary)
Feynman’s Technique (Differentiation
under the Integral Sign)
Introduce a parameter into the integral, differen-
Hints for Challenging Integrals
tiate with respect to the parameter, evaluate the
Hints for ”Hard AF” Integrals
new integral, and then integrate back with respect R√ √
Rb
to the parameter. If I(t) = a f (x, t) dx, then un- 1. ( tan x dx): Substitute u = tan x, so u2 =
Rb ∂
der certain conditions, I ′ (t) = a ∂t f (x, t) dx. tan x, 2u du = sec2 x dx = (1 + tan2 x) dx =
R 2u2
(1 + u4 ) dx. This leads to 1+u du. Factor
√ √ 4
1 + u4 = (u2 − 2u + 1)(u2 + 2u + 1) and use
Challenging Integrals partial fractions.
√
Here are some integrals known for their difficulty 2. ( x41+1 dx): Factor x4 +1 = (x2 − 2x+1)(x2 +
R
√
or requiring non-standard techniques. 2x + 1) and use partial fractions.
R x
3. ( e sin Rx dx): Use integration by parts twice.
”Hard AF” Integrals Let I = ex sin x dx. The integral will reappear
R√
1. tan x dx on the right side, allowing you to solve for I.
R 1
4. ( sec3 x dx): Use integration by parts with u =
R
2. x4 +1 dx
sec x and dv = sec2 x dx. You might need the
3. ex sin x dx (Classic by parts twice)
R
identity tan2 x = sec2 x − 1.
4. sec3 x dx
R
5. ( xx dx): This integral does not have an ele-
R
R x
5. x dx (Non-elementary) mentary antiderivative. It can be expressed as
P∞ n
R sin x an infinite series: xx = ex ln x = n=0 (x lnn!x) .
6. x dx (Si(x), non-elementary antiderivative)
R Integrating term by term is possible but yields
7. ln(ln x) dx another series.

2
 R ∞ −ax −bx
6. ( sinx x dx): The antiderivative is the Sine Inte- 4. (Frullani: 0 e −e
R
x dx): The general for-
gral function, Si(x), which is non-elementary. mula is (f (∞) − f (0)) ln(b/a) if limits f (0) and
R f (∞) exist. For f (x) = e−x , f (0) = 1, f (∞) =
7. ( ln(ln x) dx): Use integration by parts with
0. So the integral is (0 − 1) ln(b/a) = ln(a/b),
u = ln(ln x) Rand dv = dx. The resulting inte-
or often written as ln(b/a) by swapping terms in
gral will be ln1x dx, which is the Logarithmic
the numerator of the integrand. A proof involves
Integral function Li(x) (non-elementary). Ra ′
writing f (ax)−f x
(bx)
= b
f (yx)dy.
1
R
8. ( sin4 x+cos 4 x dx): Divide numerator and de- R 1 arctan x
4
nominator by cos x. This gives tan
R sec4 x 5. ( 0 x√1−x2 dx): Substitute x = sin θ. Leads to
4 x+1 dx =
R (1+tan2 x) sec2 x R π/2 arctan(sin θ)
dx. Substitute u = tan x. This 0 sin θ dθ. This is related to series ex-
tan4 x+1
R 1+u2 pansions √ and special functions. It evaluates to
becomes 1+u4 du. Divide numerator and de- π
2 ln(1 + 2).
nominator by u2 .
R∞ xs
R∞ 2
9. ( 0 e−x dx): This is the Gaussian inte- 6. ( 0 (x+a)(x+b) dx): Typically solved using con-
gral. A common tour integration by considering a keyhole con-
R trick2 is toR square2 the in-
∞ −x ∞ −y tour for z s and the poles at −a, −b. The result
tegral I 2 = 0
e dx 0
e dy =
R ∞ R ∞ −(x2 +y2 ) involves Gamma functions or Beta functions de-
π(as −bs )
0 0
e dx dy. Then√ convert to polar pending on the form. It evaluates to (a−b) sin(πs)
coordinates. The value √ is π/2. (The integral for 0 < Re(s) < 1.
from −∞ to ∞ is π). R∞
R∞ 7. ( 0 cos(ax)−cos(bx) dx): Integrate by parts first
10. ( 0 sinx x dx): This is the Dirichlet integral. Can x2
(let u = cos(ax) − cos(bx) and dv = x12 dx
be solved usingR ∞ Feynman’s technique by consid- is not ideal due to 1/x term). Better: use
ering I(t) = 0 sinx x e−tx dx and finding I ′ (t), or R ∞ f (x) R∞
using complex analysis (contour integration), or dx = − 0 f ′ (x) x1 dx if f (0) = 0, or use
R0∞ x2 R∞
(cos(ax)−cos(bx)) 0 ye−xy dy dx or relate


Laplace transforms. The value is π/2. 0
to the Dirichlet integral. The result is π(b−a)2 for
Hints for ”Almost Impossible” Inte- a, b > 0.
grals 8. (Integral from Hell ): The integral itself is di-
R1 vergent. A common regularization is to con-
1. ( 0 ln(1+x)
1+x2 dx): Substitute x = tan θ. This is a R∞ 1 
dx
π
known result: 8 ln 2. Related to ”Sophomore’s sider limϵ→0 ϵ 1+x2 − cos x x . Often in-
Dream” integrals. volves splitting the integral and using expansions
R π/2 R π/2 or known results like Euler-Mascheroni constant.
2. ( 0 ln(sin x) dx): Let I = 0 ln(sin x) dx. Value is γ (Euler-Mascheroni constant).
Ra Ra
Use the property 0 f (x) dx = 0 f (a − x) dx
R π/2 9. (Very general exponential integral ): Seldom has
to show I = 0 ln(cos x) dx. Then 2I =
R π/2 R π/2 a simple closed form unless a, b, c, d, s take very
0
ln(sin x cos x) dx = 0
ln( sin(2x)
2 ) dx. specific values. May involve special functions
This can be solved to get − π2 ln 2. like Gamma function, Mittag-Leffler function, or
R ∞ ln x
3. ( 0 (1+x2 )2 dx): Use Feynman’s technique Hypergeometric functions.
Consider I(a) = 10. ( ln1x dx): This defines the Logarithmic Inte-
R
or
R ∞ contour integration.
xa
0 (1+x2 )2
dx and then differentiate with respect gral function, Li(x), which is non-elementary.
to a and set a = 0. Alternatively, use substitu- For definite integrals, numerical methods or se-
tion x = tan θ. The value is −π/4. ries expansions are often used.