Sec on 3.1–3.2
Exponen al and Logarithmic
        Func ons
       V63.0121.001: Calculus I
     Professor Ma hew Leingang
            New York University


           March 9, 2011
                 .
Announcements

   Midterm is graded.
   average = 44, median=46,
   SD =10
   There is WebAssign due
   a er Spring Break.
   Quiz 3 on 2.6, 2.8, 3.1, 3.2
   on March 30
Midterm Statistics

   Average: 43.86/60 = 73.1%
   Median: 46/60 = 76.67%
   Standard Devia on: 10.64%
   “good” is anything above average and “great” is anything more
   than one standard devia on above average.
   More than one SD below the mean is cause for concern.
Objectives for Sections 3.1 and 3.2

    Know the defini on of an
    exponen al func on
    Know the proper es of
    exponen al func ons
    Understand and apply
    the laws of logarithms,
    including the change of
    base formula.
Outline
 Defini on of exponen al func ons

 Proper es of exponen al Func ons

 The number e and the natural exponen al func on
    Compound Interest
    The number e
    A limit

 Logarithmic Func ons
Derivation of exponentials
 Defini on
 If a is a real number and n is a posi ve whole number, then

                          an = a · a · · · · · a
                                    n factors
Derivation of exponentials
 Defini on
 If a is a real number and n is a posi ve whole number, then

                          an = a · a · · · · · a
                                    n factors



 Examples

     23 = 2 · 2 · 2 = 8
     34 = 3 · 3 · 3 · 3 = 81
     (−1)5 = (−1)(−1)(−1)(−1)(−1) = −1
Anatomy of a power

 Defini on
 A power is an expression of the form ab .
     The number a is called the base.
     The number b is called the exponent.
Fact
If a is a real number, then
       ax+y = ax ay (sums to products)
        x−y    ax
       a = y
               a
       (ax )y = axy
       (ab)x = ax bx
whenever all exponents are posi ve whole numbers.
Fact
If a is a real number, then
       ax+y = ax ay (sums to products)
        x−y    ax
       a = y (differences to quo ents)
               a
       (ax )y = axy
       (ab)x = ax bx
whenever all exponents are posi ve whole numbers.
Fact
If a is a real number, then
       ax+y = ax ay (sums to products)
        x−y    ax
       a = y (differences to quo ents)
               a
       (ax )y = axy (repeated exponen a on to mul plied powers)
       (ab)x = ax bx
whenever all exponents are posi ve whole numbers.
Fact
If a is a real number, then
       ax+y = ax ay (sums to products)
        x−y    ax
       a = y (differences to quo ents)
               a
       (ax )y = axy (repeated exponen a on to mul plied powers)
       (ab)x = ax bx (power of product is product of powers)
whenever all exponents are posi ve whole numbers.
Fact
If a is a real number, then
       ax+y = ax ay (sums to products)
        x−y    ax
       a = y (differences to quo ents)
               a
       (ax )y = axy (repeated exponen a on to mul plied powers)
       (ab)x = ax bx (power of product is product of powers)
whenever all exponents are posi ve whole numbers.
Fact
If a is a real number, then
       ax+y = ax ay (sums to products)
        x−y    ax
       a = y (differences to quo ents)
               a
       (ax )y = axy (repeated exponen a on to mul plied powers)
       (ab)x = ax bx (power of product is product of powers)
whenever all exponents are posi ve whole numbers.

Proof.
Check for yourself:

         ax+y = a · a · · · · · a = a · a · · · · · a · a · a · · · · · a = ax ay
                    x + y factors         x factors        y factors
Let’s be conventional
   The desire that these proper es remain true gives us
   conven ons for ax when x is not a posi ve whole number.
Let’s be conventional
   The desire that these proper es remain true gives us
   conven ons for ax when x is not a posi ve whole number.
   For example, what should a0 be?
   We would want this to be true:
                         !
              an = an+0 = an · a0
Let’s be conventional
   The desire that these proper es remain true gives us
   conven ons for ax when x is not a posi ve whole number.
   For example, what should a0 be?
   We would want this to be true:
                                                n
                           !                ! a
               an = an+0 = an · a0 =⇒ a0 = n = 1
                                              a
Let’s be conventional
     The desire that these proper es remain true gives us
     conven ons for ax when x is not a posi ve whole number.
     For example, what should a0 be?
     We would want this to be true:
                                                  n
                             !                ! a
                 an = an+0 = an · a0 =⇒ a0 = n = 1
                                                a

 Defini on
 If a ̸= 0, we define a0 = 1.
Let’s be conventional
     The desire that these proper es remain true gives us
     conven ons for ax when x is not a posi ve whole number.
     For example, what should a0 be?
     We would want this to be true:
                                                  n
                             !                ! a
                 an = an+0 = an · a0 =⇒ a0 = n = 1
                                                a

 Defini on
 If a ̸= 0, we define a0 = 1.

     No ce 00 remains undefined (as a limit form, it’s
     indeterminate).
Conventions for negative exponents

 If n ≥ 0, we want

              an+(−n) = an · a−n
                      !
Conventions for negative exponents

 If n ≥ 0, we want
                                            a0   1
              an+(−n) = an · a−n =⇒ a−n =
                     !                 !
                                               = n
                                            an  a
Conventions for negative exponents

 If n ≥ 0, we want
                                                    a0   1
               an+(−n) = an · a−n =⇒ a−n =
                       !                       !
                                                       = n
                                                    an  a

 Defini on
                                             1
 If n is a posi ve integer, we define a−n =      .
                                             an
Defini on
                                            1
If n is a posi ve integer, we define a−n =      .
                                            an
Defini on
                                            1
If n is a posi ve integer, we define a−n =      .
                                            an

Fact
                                    1
       The conven on that a−n =       “works” for nega ve n as well.
                                   an
                                           m−n  am
       If m and n are any integers, then a     = n.
                                                 a
Conventions for fractional exponents
 If q is a posi ve integer, we want
                           !
                  (a1/q )q = a1 = a
Conventions for fractional exponents
 If q is a posi ve integer, we want
                           !               !    √
                  (a1/q )q = a1 = a =⇒ a1/q =   q
                                                    a
Conventions for fractional exponents
 If q is a posi ve integer, we want
                           !                     !   √
                  (a1/q )q = a1 = a =⇒ a1/q =        q
                                                         a


 Defini on
                                              √
 If q is a posi ve integer, we define a1/q =   q
                                                a. We must have a ≥ 0
 if q is even.
Conventions for fractional exponents
 If q is a posi ve integer, we want
                           !                     !   √
                  (a1/q )q = a1 = a =⇒ a1/q =        q
                                                         a


 Defini on
                                           √
 If q is a posi ve integer, we define a1/q = q a. We must have a ≥ 0
 if q is even.
               √q
                       (√ )p
 No ce that ap = q a . So we can unambiguously say

                        ap/q = (ap )1/q = (a1/q )p
Conventions for irrational
exponents
   So ax is well-defined if a is posi ve and x is ra onal.
   What about irra onal powers?
Conventions for irrational
exponents
     So ax is well-defined if a is posi ve and x is ra onal.
     What about irra onal powers?

 Defini on
 Let a > 0. Then
                            ax = lim ar
                                  r→x
                                  r ra onal
Conventions for irrational
exponents
      So ax is well-defined if a is posi ve and x is ra onal.
      What about irra onal powers?

 Defini on
 Let a > 0. Then
                             ax = lim ar
                                   r→x
                                   r ra onal


 In other words, to approximate ax for irra onal x, take r close to x
 but ra onal and compute ar .
Approximating a power with an
irrational exponent
                 r                    2r
                 3                 23
                                   √=8
                                   10
                3.1     231/10 = √ 31 ≈ 8.57419
                                      2
                                  100
               3.14    2314/100 = √ 314 ≈ 8.81524
                                       2
                                   1000
              3.141 23141/1000 =          23141 ≈ 8.82135
 The limit (numerically approximated is)

                            2π ≈ 8.82498
Graphs of exponential functions
                y




                .           x
Graphs of exponential functions
                y




                            y = 1x
                .           x
Graphs of exponential functions
                y
                       y = 2x




                                y = 1x
                .               x
Graphs of exponential functions
                y
                     y = 3x = 2x
                          y




                                   y = 1x
                .                  x
Graphs of exponential functions
                y
                    y = 10x 3x = 2x
                        y= y




                                      y = 1x
                .                     x
Graphs of exponential functions
                y
                    y = 10x 3x = 2x
                        y= y          y = 1.5x




                                      y = 1x
                .                     x
Graphs of exponential functions
                   y
      y = (1/2)x       y = 10x 3x = 2x
                           y= y          y = 1.5x




                                         y = 1x
                   .                     x
Graphs of exponential functions
                         y
      y = (y/= x(1/3)x
           1 2)              y = 10x 3x = 2x
                                 y= y          y = 1.5x




                                               y = 1x
                         .                     x
Graphs of exponential functions
                           y
      y = (y/= x(1/3)x
           1 2)          y = (1/10y x= 10x 3x = 2x
                                  ) y= y             y = 1.5x




                                                     y = 1x
                            .                        x
Graphs of exponential functions
                          y
      y y =y/=3(1/3)x
        = (1(2/x)x
             2)         y = (1/10y x= 10x 3x = 2x
                                 ) y= y             y = 1.5x




                                                    y = 1x
                           .                        x
Outline
 Defini on of exponen al func ons

 Proper es of exponen al Func ons

 The number e and the natural exponen al func on
    Compound Interest
    The number e
    A limit

 Logarithmic Func ons
Properties of exponential Functions
 Theorem
 If a > 0 and a ̸= 1, then f(x) = ax is a con nuous func on with
 domain (−∞, ∞) and range (0, ∞). In par cular, ax > 0 for all x.
 For any real numbers x and y, and posi ve numbers a and b we have
       ax+y = ax ay
        x−y   ax
       a = y
              a
       (a ) = axy
         x y

       (ab)x = ax bx
Properties of exponential Functions
 Theorem
 If a > 0 and a ̸= 1, then f(x) = ax is a con nuous func on with
 domain (−∞, ∞) and range (0, ∞). In par cular, ax > 0 for all x.
 For any real numbers x and y, and posi ve numbers a and b we have
       ax+y = ax ay
        x−y   ax
       a = y (nega ve exponents mean reciprocals)
              a
       (a ) = axy
         x y

       (ab)x = ax bx
Properties of exponential Functions
 Theorem
 If a > 0 and a ̸= 1, then f(x) = ax is a con nuous func on with
 domain (−∞, ∞) and range (0, ∞). In par cular, ax > 0 for all x.
 For any real numbers x and y, and posi ve numbers a and b we have
       ax+y = ax ay
        x−y   ax
       a = y (nega ve exponents mean reciprocals)
              a
       (a ) = axy (frac onal exponents mean roots)
         x y

       (ab)x = ax bx
Proof.
    This is true for posi ve integer exponents by natural defini on
    Our conven onal defini ons make these true for ra onal
    exponents
    Our limit defini on make these for irra onal exponents, too
Simplifying exponential expressions

 Example
 Simplify: 82/3
Simplifying exponential expressions

 Example
 Simplify: 82/3

 Solu on
              √
              3
                   √
       2/3
             = 82 = 64 = 4
                    3
      8
Simplifying exponential expressions

 Example
 Simplify: 82/3

 Solu on
             √3
                  √
       2/3
      8 = 82 = 64 = 4
                   3

          (√ )2
             8 = 22 = 4.
           3
      Or,
Simplifying exponential
expressions

 Example
             √
                 8
 Simplify:
             21/2
Simplifying exponential
expressions

 Example
             √
                 8
 Simplify:
             21/2

 Answer
 2
Limits of exponential functions
Fact (Limits of exponen al
func ons)                                       y
                               y (1 y )/3 x
                             y = =/(1= )(2/3)x y = y1/1010= 2x = 1.5
                                    2
                                       x
                                                   ( = =x 3x y
                                                     y ) x
                                                        y
    If a > 1, then
     lim ax = ∞ and
    x→∞
     lim ax = 0
    x→−∞
    If 0 < a < 1, then                                      y = 1x
     lim ax = 0 and                            .            x
    x→∞
     lim ax = ∞
    x→−∞
Outline
 Defini on of exponen al func ons

 Proper es of exponen al Func ons

 The number e and the natural exponen al func on
    Compound Interest
    The number e
    A limit

 Logarithmic Func ons
Compounded Interest
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded once a year. How much do you have A er one year?
 A er two years? A er t years?
Compounded Interest
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded once a year. How much do you have A er one year?
 A er two years? A er t years?

 Answer
     $100 + 10% = $110
Compounded Interest
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded once a year. How much do you have A er one year?
 A er two years? A er t years?

 Answer
     $100 + 10% = $110
     $110 + 10% = $110 + $11 = $121
Compounded Interest
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded once a year. How much do you have A er one year?
 A er two years? A er t years?

 Answer
     $100 + 10% = $110
     $110 + 10% = $110 + $11 = $121
     $100(1.1)t .
Compounded Interest: quarterly
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded four mes a year. How much do you have A er one
 year? A er two years? A er t years?
Compounded Interest: quarterly
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded four mes a year. How much do you have A er one
 year? A er two years? A er t years?

 Answer
     $100(1.025)4 = $110.38,
Compounded Interest: quarterly
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded four mes a year. How much do you have A er one
 year? A er two years? A er t years?

 Answer
     $100(1.025)4 = $110.38, not $100(1.1)4 !
Compounded Interest: quarterly
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded four mes a year. How much do you have A er one
 year? A er two years? A er t years?

 Answer
     $100(1.025)4 = $110.38, not $100(1.1)4 !
     $100(1.025)8 = $121.84
Compounded Interest: quarterly
 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded four mes a year. How much do you have A er one
 year? A er two years? A er t years?

 Answer
     $100(1.025)4 = $110.38, not $100(1.1)4 !
     $100(1.025)8 = $121.84
     $100(1.025)4t .
Compounded Interest: monthly

 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded twelve mes a year. How much do you have a er t
 years?
Compounded Interest: monthly

 Ques on
 Suppose you save $100 at 10% annual interest, with interest
 compounded twelve mes a year. How much do you have a er t
 years?

 Answer
 $100(1 + 10%/12)12t
Compounded Interest: general
 Ques on
 Suppose you save P at interest rate r, with interest compounded n
  mes a year. How much do you have a er t years?
Compounded Interest: general
 Ques on
 Suppose you save P at interest rate r, with interest compounded n
  mes a year. How much do you have a er t years?

 Answer
                                 (    r )nt
                         B(t) = P 1 +
                                      n
Compounded Interest: continuous
 Ques on
 Suppose you save P at interest rate r, with interest compounded
 every instant. How much do you have a er t years?
Compounded Interest: continuous
 Ques on
 Suppose you save P at interest rate r, with interest compounded
 every instant. How much do you have a er t years?

 Answer
                         (                (      )rnt
                             r )nt             1
            B(t) = lim P 1 +       = lim P 1 +
                   n→∞       n       n→∞       n
                     [     (      )n ]rt
                                1
                 = P lim 1 +
                       n→∞      n
                      independent of P, r, or t
The magic number
 Defini on
                    (     )n
                        1
            e = lim 1 +
                n→∞     n
The magic number
 Defini on
                               (     )n
                                   1
                       e = lim 1 +
                           n→∞     n

 So now con nuously-compounded interest can be expressed as

                          B(t) = Pert .
Existence of e
See Appendix B
                     (      )n
                          1
                 n     1+
                          n
                 1   2
                 2   2.25
Existence of e
See Appendix B
                     (      )n
                          1
                 n     1+
                          n
                 1   2
                 2   2.25
                 3   2.37037
Existence of e
See Appendix B
                      (      )n
                           1
                 n      1+
                           n
                 1    2
                 2    2.25
                 3    2.37037
                 10   2.59374
Existence of e
See Appendix B
                       (      )n
                            1
                 n       1+
                            n
                 1     2
                 2     2.25
                 3     2.37037
                 10    2.59374
                 100   2.70481
Existence of e
See Appendix B
                      (      )n
                           1
                 n      1+
                           n
                 1    2
                 2    2.25
                 3    2.37037
                 10   2.59374
                 100 2.70481
                 1000 2.71692
Existence of e
See Appendix B
                      (      )n
                           1
                 n      1+
                           n
                 1    2
                 2    2.25
                 3    2.37037
                 10   2.59374
                 100 2.70481
                 1000 2.71692
                 106 2.71828
Existence of e
See Appendix B
                                      (      )n
                                           1
                                 n      1+
   We can experimentally                   n
   verify that this number       1    2
   exists and is                 2    2.25
                                 3    2.37037
   e ≈ 2.718281828459045 . . .
                                 10   2.59374
                                 100 2.70481
                                 1000 2.71692
                                 106 2.71828
Existence of e
See Appendix B
                                      (      )n
                                           1
                                 n      1+
   We can experimentally                   n
   verify that this number       1    2
   exists and is                 2    2.25
                                 3    2.37037
   e ≈ 2.718281828459045 . . .
                                 10   2.59374
   e is irra onal                100 2.70481
                                 1000 2.71692
                                 106 2.71828
Existence of e
See Appendix B
                                      (      )n
                                           1
                                 n      1+
   We can experimentally                   n
   verify that this number       1    2
   exists and is                 2    2.25
                                 3    2.37037
   e ≈ 2.718281828459045 . . .
                                 10   2.59374
   e is irra onal                100 2.70481
                                 1000 2.71692
   e is transcendental
                                 106 2.71828
Meet the Mathematician: Leonhard Euler
   Born in Switzerland, lived
   in Prussia (Germany) and
   Russia
   Eyesight trouble all his
   life, blind from 1766
   onward
   Hundreds of
   contribu ons to calculus,
   number theory, graph
   theory, fluid mechanics,      Leonhard Paul Euler
   op cs, and astronomy          Swiss, 1707–1783
A limit
 Ques on
             eh − 1
 What is lim        ?
         h→0    h
A limit
 Ques on
             eh − 1
 What is lim        ?
         h→0    h
 Answer
     e = lim (1 + 1/n)n = lim (1 + h)1/h . So for a small h,
          n→∞                 h→0
     e ≈ (1 + h) 1/h
                       . So
                                 [           ]h
                        eh − 1     (1 + h)1/h − 1
                               ≈                  =1
                           h              h
A limit

                       eh − 1
    It follows that lim       = 1.
                   h→0    h
                                            2h − 1
    This can be used to characterize e: lim        = 0.693 · · · < 1
                                        h→0    h
             3h − 1
    and lim         = 1.099 · · · > 1
         h→0    h
Outline
 Defini on of exponen al func ons

 Proper es of exponen al Func ons

 The number e and the natural exponen al func on
    Compound Interest
    The number e
    A limit

 Logarithmic Func ons
Logarithms
 Defini on
    The base a logarithm loga x is the inverse of the func on ax

                         y = loga x ⇐⇒ x = ay

    The natural logarithm ln x is the inverse of ex . So
    y = ln x ⇐⇒ x = ey .
Facts about Logarithms

 Facts
  (i) loga (x1 · x2 ) = loga x1 + loga x2
Facts about Logarithms

 Facts
  (i) loga (x1 · x2 ) = loga x1 + loga x2
           ( )
             x1
 (ii) loga          = loga x1 − loga x2
             x2
Facts about Logarithms

 Facts
   (i) loga (x1 · x2 ) = loga x1 + loga x2
            ( )
               x1
  (ii) loga          = loga x1 − loga x2
               x2
 (iii) loga (xr ) = r loga x
Logarithms convert products to sums
    Suppose y1 = loga x1 and y2 = loga x2
    Then x1 = ay1 and x2 = ay2
    So x1 x2 = ay1 ay2 = ay1 +y2
    Therefore
                       loga (x1 · x2 ) = loga x1 + loga x2
Examples
 Example
 Write as a single logarithm: 2 ln 4 − ln 3.
Examples
 Example
 Write as a single logarithm: 2 ln 4 − ln 3.

 Solu on
                                          42
      2 ln 4 − ln 3 = ln 42 − ln 3 = ln
                                          3
            ln 42
      not         !
             ln 3
Examples
 Example
                                   3
 Write as a single logarithm: ln     + 4 ln 2
                                   4
Examples
 Example
                                   3
 Write as a single logarithm: ln     + 4 ln 2
                                   4

 Solu on

           3
      ln     + 4 ln 2 = ln 3 − ln 4 + 4 ln 2 = ln 3 − 2 ln 2 + 4 ln 2
           4
                      = ln 3 + 2 ln 2 = ln(3 · 22 ) = ln 12
Graphs of logarithmic functions
       y
                    y = 2x

                             y = log2 x


           (0, 1)
        .
      (1, 0)                      x
Graphs of logarithmic functions
       y
                    y =y3= 2x
                         x



                                y = log2 x

                                y = log3 x
           (0, 1)
        .
      (1, 0)                         x
Graphs of logarithmic functions
       y
               y =y10y3= 2x
                    =x x

                              y = log2 x

                              y = log3 x
           (0, 1)
                              y = log10 x
        .
      (1, 0)                        x
Graphs of logarithmic functions
       y
                   y =x ex
               y =y10y3= 2x
                     = x

                              y = log2 x

                              yy= log3 x
                                 = ln x
           (0, 1)
                              y = log10 x
        .
      (1, 0)                        x
Change of base formula for logarithms
 Fact
                                                          logb x
 If a > 0 and a ̸= 1, and the same for b, then loga x =
                                                          logb a
Change of base formula for logarithms
 Fact
                                                          logb x
 If a > 0 and a ̸= 1, and the same for b, then loga x =
                                                          logb a

 Proof.
        If y = loga x, then x = ay
        So logb x = logb (ay ) = y logb a
        Therefore
                                             logb x
                                y = loga x =
                                             logb a
Example of changing base

 Example
 Find log2 8 by using log10 only.
Example of changing base

 Example
 Find log2 8 by using log10 only.

 Solu on
            log10 8 0.90309
 log2 8 =          ≈        =3
            log10 2 0.30103
Example of changing base

 Example
 Find log2 8 by using log10 only.

 Solu on
            log10 8 0.90309
 log2 8 =          ≈        =3
            log10 2 0.30103
 Surprised?
Example of changing base

 Example
 Find log2 8 by using log10 only.

 Solu on
            log10 8 0.90309
 log2 8 =          ≈        =3
            log10 2 0.30103
 Surprised? No, log2 8 = log2 23 = 3 directly.
Upshot of changing base
 The point of the change of base formula
                    logb x   1
         loga x =          =      · logb x = constant · logb x
                    logb a logb a
 is that all the logarithmic func ons are mul ples of each other. So
 just pick one and call it your favorite.
       Engineers like the common logarithm log = log10
       Computer scien sts like the binary logarithm lg = log2
       Mathema cians like natural logarithm ln = loge
 Naturally, we will follow the mathema cians. Just don’t pronounce
 it “lawn.”
“lawn”
                         .




        .


Image credit: Selva
Summary


  Exponen als turn sums into products
  Logarithms turn products into sums
  Slide rule scabbards are wicked cool

Lesson 13: Exponential and Logarithmic Functions (slides)

  • 1.
    Sec on 3.1–3.2 Exponenal and Logarithmic Func ons V63.0121.001: Calculus I Professor Ma hew Leingang New York University March 9, 2011 .
  • 2.
    Announcements Midterm is graded. average = 44, median=46, SD =10 There is WebAssign due a er Spring Break. Quiz 3 on 2.6, 2.8, 3.1, 3.2 on March 30
  • 3.
    Midterm Statistics Average: 43.86/60 = 73.1% Median: 46/60 = 76.67% Standard Devia on: 10.64% “good” is anything above average and “great” is anything more than one standard devia on above average. More than one SD below the mean is cause for concern.
  • 4.
    Objectives for Sections3.1 and 3.2 Know the defini on of an exponen al func on Know the proper es of exponen al func ons Understand and apply the laws of logarithms, including the change of base formula.
  • 5.
    Outline Defini onof exponen al func ons Proper es of exponen al Func ons The number e and the natural exponen al func on Compound Interest The number e A limit Logarithmic Func ons
  • 6.
    Derivation of exponentials Defini on If a is a real number and n is a posi ve whole number, then an = a · a · · · · · a n factors
  • 7.
    Derivation of exponentials Defini on If a is a real number and n is a posi ve whole number, then an = a · a · · · · · a n factors Examples 23 = 2 · 2 · 2 = 8 34 = 3 · 3 · 3 · 3 = 81 (−1)5 = (−1)(−1)(−1)(−1)(−1) = −1
  • 8.
    Anatomy of apower Defini on A power is an expression of the form ab . The number a is called the base. The number b is called the exponent.
  • 9.
    Fact If a isa real number, then ax+y = ax ay (sums to products) x−y ax a = y a (ax )y = axy (ab)x = ax bx whenever all exponents are posi ve whole numbers.
  • 10.
    Fact If a isa real number, then ax+y = ax ay (sums to products) x−y ax a = y (differences to quo ents) a (ax )y = axy (ab)x = ax bx whenever all exponents are posi ve whole numbers.
  • 11.
    Fact If a isa real number, then ax+y = ax ay (sums to products) x−y ax a = y (differences to quo ents) a (ax )y = axy (repeated exponen a on to mul plied powers) (ab)x = ax bx whenever all exponents are posi ve whole numbers.
  • 12.
    Fact If a isa real number, then ax+y = ax ay (sums to products) x−y ax a = y (differences to quo ents) a (ax )y = axy (repeated exponen a on to mul plied powers) (ab)x = ax bx (power of product is product of powers) whenever all exponents are posi ve whole numbers.
  • 13.
    Fact If a isa real number, then ax+y = ax ay (sums to products) x−y ax a = y (differences to quo ents) a (ax )y = axy (repeated exponen a on to mul plied powers) (ab)x = ax bx (power of product is product of powers) whenever all exponents are posi ve whole numbers.
  • 14.
    Fact If a isa real number, then ax+y = ax ay (sums to products) x−y ax a = y (differences to quo ents) a (ax )y = axy (repeated exponen a on to mul plied powers) (ab)x = ax bx (power of product is product of powers) whenever all exponents are posi ve whole numbers. Proof. Check for yourself: ax+y = a · a · · · · · a = a · a · · · · · a · a · a · · · · · a = ax ay x + y factors x factors y factors
  • 15.
    Let’s be conventional The desire that these proper es remain true gives us conven ons for ax when x is not a posi ve whole number.
  • 16.
    Let’s be conventional The desire that these proper es remain true gives us conven ons for ax when x is not a posi ve whole number. For example, what should a0 be? We would want this to be true: ! an = an+0 = an · a0
  • 17.
    Let’s be conventional The desire that these proper es remain true gives us conven ons for ax when x is not a posi ve whole number. For example, what should a0 be? We would want this to be true: n ! ! a an = an+0 = an · a0 =⇒ a0 = n = 1 a
  • 18.
    Let’s be conventional The desire that these proper es remain true gives us conven ons for ax when x is not a posi ve whole number. For example, what should a0 be? We would want this to be true: n ! ! a an = an+0 = an · a0 =⇒ a0 = n = 1 a Defini on If a ̸= 0, we define a0 = 1.
  • 19.
    Let’s be conventional The desire that these proper es remain true gives us conven ons for ax when x is not a posi ve whole number. For example, what should a0 be? We would want this to be true: n ! ! a an = an+0 = an · a0 =⇒ a0 = n = 1 a Defini on If a ̸= 0, we define a0 = 1. No ce 00 remains undefined (as a limit form, it’s indeterminate).
  • 20.
    Conventions for negativeexponents If n ≥ 0, we want an+(−n) = an · a−n !
  • 21.
    Conventions for negativeexponents If n ≥ 0, we want a0 1 an+(−n) = an · a−n =⇒ a−n = ! ! = n an a
  • 22.
    Conventions for negativeexponents If n ≥ 0, we want a0 1 an+(−n) = an · a−n =⇒ a−n = ! ! = n an a Defini on 1 If n is a posi ve integer, we define a−n = . an
  • 23.
    Defini on 1 If n is a posi ve integer, we define a−n = . an
  • 24.
    Defini on 1 If n is a posi ve integer, we define a−n = . an Fact 1 The conven on that a−n = “works” for nega ve n as well. an m−n am If m and n are any integers, then a = n. a
  • 25.
    Conventions for fractionalexponents If q is a posi ve integer, we want ! (a1/q )q = a1 = a
  • 26.
    Conventions for fractionalexponents If q is a posi ve integer, we want ! ! √ (a1/q )q = a1 = a =⇒ a1/q = q a
  • 27.
    Conventions for fractionalexponents If q is a posi ve integer, we want ! ! √ (a1/q )q = a1 = a =⇒ a1/q = q a Defini on √ If q is a posi ve integer, we define a1/q = q a. We must have a ≥ 0 if q is even.
  • 28.
    Conventions for fractionalexponents If q is a posi ve integer, we want ! ! √ (a1/q )q = a1 = a =⇒ a1/q = q a Defini on √ If q is a posi ve integer, we define a1/q = q a. We must have a ≥ 0 if q is even. √q (√ )p No ce that ap = q a . So we can unambiguously say ap/q = (ap )1/q = (a1/q )p
  • 29.
    Conventions for irrational exponents So ax is well-defined if a is posi ve and x is ra onal. What about irra onal powers?
  • 30.
    Conventions for irrational exponents So ax is well-defined if a is posi ve and x is ra onal. What about irra onal powers? Defini on Let a > 0. Then ax = lim ar r→x r ra onal
  • 31.
    Conventions for irrational exponents So ax is well-defined if a is posi ve and x is ra onal. What about irra onal powers? Defini on Let a > 0. Then ax = lim ar r→x r ra onal In other words, to approximate ax for irra onal x, take r close to x but ra onal and compute ar .
  • 32.
    Approximating a powerwith an irrational exponent r 2r 3 23 √=8 10 3.1 231/10 = √ 31 ≈ 8.57419 2 100 3.14 2314/100 = √ 314 ≈ 8.81524 2 1000 3.141 23141/1000 = 23141 ≈ 8.82135 The limit (numerically approximated is) 2π ≈ 8.82498
  • 33.
    Graphs of exponentialfunctions y . x
  • 34.
    Graphs of exponentialfunctions y y = 1x . x
  • 35.
    Graphs of exponentialfunctions y y = 2x y = 1x . x
  • 36.
    Graphs of exponentialfunctions y y = 3x = 2x y y = 1x . x
  • 37.
    Graphs of exponentialfunctions y y = 10x 3x = 2x y= y y = 1x . x
  • 38.
    Graphs of exponentialfunctions y y = 10x 3x = 2x y= y y = 1.5x y = 1x . x
  • 39.
    Graphs of exponentialfunctions y y = (1/2)x y = 10x 3x = 2x y= y y = 1.5x y = 1x . x
  • 40.
    Graphs of exponentialfunctions y y = (y/= x(1/3)x 1 2) y = 10x 3x = 2x y= y y = 1.5x y = 1x . x
  • 41.
    Graphs of exponentialfunctions y y = (y/= x(1/3)x 1 2) y = (1/10y x= 10x 3x = 2x ) y= y y = 1.5x y = 1x . x
  • 42.
    Graphs of exponentialfunctions y y y =y/=3(1/3)x = (1(2/x)x 2) y = (1/10y x= 10x 3x = 2x ) y= y y = 1.5x y = 1x . x
  • 43.
    Outline Defini onof exponen al func ons Proper es of exponen al Func ons The number e and the natural exponen al func on Compound Interest The number e A limit Logarithmic Func ons
  • 44.
    Properties of exponentialFunctions Theorem If a > 0 and a ̸= 1, then f(x) = ax is a con nuous func on with domain (−∞, ∞) and range (0, ∞). In par cular, ax > 0 for all x. For any real numbers x and y, and posi ve numbers a and b we have ax+y = ax ay x−y ax a = y a (a ) = axy x y (ab)x = ax bx
  • 45.
    Properties of exponentialFunctions Theorem If a > 0 and a ̸= 1, then f(x) = ax is a con nuous func on with domain (−∞, ∞) and range (0, ∞). In par cular, ax > 0 for all x. For any real numbers x and y, and posi ve numbers a and b we have ax+y = ax ay x−y ax a = y (nega ve exponents mean reciprocals) a (a ) = axy x y (ab)x = ax bx
  • 46.
    Properties of exponentialFunctions Theorem If a > 0 and a ̸= 1, then f(x) = ax is a con nuous func on with domain (−∞, ∞) and range (0, ∞). In par cular, ax > 0 for all x. For any real numbers x and y, and posi ve numbers a and b we have ax+y = ax ay x−y ax a = y (nega ve exponents mean reciprocals) a (a ) = axy (frac onal exponents mean roots) x y (ab)x = ax bx
  • 47.
    Proof. This is true for posi ve integer exponents by natural defini on Our conven onal defini ons make these true for ra onal exponents Our limit defini on make these for irra onal exponents, too
  • 48.
  • 49.
    Simplifying exponential expressions Example Simplify: 82/3 Solu on √ 3 √ 2/3 = 82 = 64 = 4 3 8
  • 50.
    Simplifying exponential expressions Example Simplify: 82/3 Solu on √3 √ 2/3 8 = 82 = 64 = 4 3 (√ )2 8 = 22 = 4. 3 Or,
  • 51.
  • 52.
    Simplifying exponential expressions Example √ 8 Simplify: 21/2 Answer 2
  • 53.
    Limits of exponentialfunctions Fact (Limits of exponen al func ons) y y (1 y )/3 x y = =/(1= )(2/3)x y = y1/1010= 2x = 1.5 2 x ( = =x 3x y y ) x y If a > 1, then lim ax = ∞ and x→∞ lim ax = 0 x→−∞ If 0 < a < 1, then y = 1x lim ax = 0 and . x x→∞ lim ax = ∞ x→−∞
  • 54.
    Outline Defini onof exponen al func ons Proper es of exponen al Func ons The number e and the natural exponen al func on Compound Interest The number e A limit Logarithmic Func ons
  • 55.
    Compounded Interest Queson Suppose you save $100 at 10% annual interest, with interest compounded once a year. How much do you have A er one year? A er two years? A er t years?
  • 56.
    Compounded Interest Queson Suppose you save $100 at 10% annual interest, with interest compounded once a year. How much do you have A er one year? A er two years? A er t years? Answer $100 + 10% = $110
  • 57.
    Compounded Interest Queson Suppose you save $100 at 10% annual interest, with interest compounded once a year. How much do you have A er one year? A er two years? A er t years? Answer $100 + 10% = $110 $110 + 10% = $110 + $11 = $121
  • 58.
    Compounded Interest Queson Suppose you save $100 at 10% annual interest, with interest compounded once a year. How much do you have A er one year? A er two years? A er t years? Answer $100 + 10% = $110 $110 + 10% = $110 + $11 = $121 $100(1.1)t .
  • 59.
    Compounded Interest: quarterly Ques on Suppose you save $100 at 10% annual interest, with interest compounded four mes a year. How much do you have A er one year? A er two years? A er t years?
  • 60.
    Compounded Interest: quarterly Ques on Suppose you save $100 at 10% annual interest, with interest compounded four mes a year. How much do you have A er one year? A er two years? A er t years? Answer $100(1.025)4 = $110.38,
  • 61.
    Compounded Interest: quarterly Ques on Suppose you save $100 at 10% annual interest, with interest compounded four mes a year. How much do you have A er one year? A er two years? A er t years? Answer $100(1.025)4 = $110.38, not $100(1.1)4 !
  • 62.
    Compounded Interest: quarterly Ques on Suppose you save $100 at 10% annual interest, with interest compounded four mes a year. How much do you have A er one year? A er two years? A er t years? Answer $100(1.025)4 = $110.38, not $100(1.1)4 ! $100(1.025)8 = $121.84
  • 63.
    Compounded Interest: quarterly Ques on Suppose you save $100 at 10% annual interest, with interest compounded four mes a year. How much do you have A er one year? A er two years? A er t years? Answer $100(1.025)4 = $110.38, not $100(1.1)4 ! $100(1.025)8 = $121.84 $100(1.025)4t .
  • 64.
    Compounded Interest: monthly Ques on Suppose you save $100 at 10% annual interest, with interest compounded twelve mes a year. How much do you have a er t years?
  • 65.
    Compounded Interest: monthly Ques on Suppose you save $100 at 10% annual interest, with interest compounded twelve mes a year. How much do you have a er t years? Answer $100(1 + 10%/12)12t
  • 66.
    Compounded Interest: general Ques on Suppose you save P at interest rate r, with interest compounded n mes a year. How much do you have a er t years?
  • 67.
    Compounded Interest: general Ques on Suppose you save P at interest rate r, with interest compounded n mes a year. How much do you have a er t years? Answer ( r )nt B(t) = P 1 + n
  • 68.
    Compounded Interest: continuous Ques on Suppose you save P at interest rate r, with interest compounded every instant. How much do you have a er t years?
  • 69.
    Compounded Interest: continuous Ques on Suppose you save P at interest rate r, with interest compounded every instant. How much do you have a er t years? Answer ( ( )rnt r )nt 1 B(t) = lim P 1 + = lim P 1 + n→∞ n n→∞ n [ ( )n ]rt 1 = P lim 1 + n→∞ n independent of P, r, or t
  • 70.
    The magic number Defini on ( )n 1 e = lim 1 + n→∞ n
  • 71.
    The magic number Defini on ( )n 1 e = lim 1 + n→∞ n So now con nuously-compounded interest can be expressed as B(t) = Pert .
  • 72.
    Existence of e SeeAppendix B ( )n 1 n 1+ n 1 2 2 2.25
  • 73.
    Existence of e SeeAppendix B ( )n 1 n 1+ n 1 2 2 2.25 3 2.37037
  • 74.
    Existence of e SeeAppendix B ( )n 1 n 1+ n 1 2 2 2.25 3 2.37037 10 2.59374
  • 75.
    Existence of e SeeAppendix B ( )n 1 n 1+ n 1 2 2 2.25 3 2.37037 10 2.59374 100 2.70481
  • 76.
    Existence of e SeeAppendix B ( )n 1 n 1+ n 1 2 2 2.25 3 2.37037 10 2.59374 100 2.70481 1000 2.71692
  • 77.
    Existence of e SeeAppendix B ( )n 1 n 1+ n 1 2 2 2.25 3 2.37037 10 2.59374 100 2.70481 1000 2.71692 106 2.71828
  • 78.
    Existence of e SeeAppendix B ( )n 1 n 1+ We can experimentally n verify that this number 1 2 exists and is 2 2.25 3 2.37037 e ≈ 2.718281828459045 . . . 10 2.59374 100 2.70481 1000 2.71692 106 2.71828
  • 79.
    Existence of e SeeAppendix B ( )n 1 n 1+ We can experimentally n verify that this number 1 2 exists and is 2 2.25 3 2.37037 e ≈ 2.718281828459045 . . . 10 2.59374 e is irra onal 100 2.70481 1000 2.71692 106 2.71828
  • 80.
    Existence of e SeeAppendix B ( )n 1 n 1+ We can experimentally n verify that this number 1 2 exists and is 2 2.25 3 2.37037 e ≈ 2.718281828459045 . . . 10 2.59374 e is irra onal 100 2.70481 1000 2.71692 e is transcendental 106 2.71828
  • 81.
    Meet the Mathematician:Leonhard Euler Born in Switzerland, lived in Prussia (Germany) and Russia Eyesight trouble all his life, blind from 1766 onward Hundreds of contribu ons to calculus, number theory, graph theory, fluid mechanics, Leonhard Paul Euler op cs, and astronomy Swiss, 1707–1783
  • 82.
    A limit Queson eh − 1 What is lim ? h→0 h
  • 83.
    A limit Queson eh − 1 What is lim ? h→0 h Answer e = lim (1 + 1/n)n = lim (1 + h)1/h . So for a small h, n→∞ h→0 e ≈ (1 + h) 1/h . So [ ]h eh − 1 (1 + h)1/h − 1 ≈ =1 h h
  • 84.
    A limit eh − 1 It follows that lim = 1. h→0 h 2h − 1 This can be used to characterize e: lim = 0.693 · · · < 1 h→0 h 3h − 1 and lim = 1.099 · · · > 1 h→0 h
  • 85.
    Outline Defini onof exponen al func ons Proper es of exponen al Func ons The number e and the natural exponen al func on Compound Interest The number e A limit Logarithmic Func ons
  • 86.
    Logarithms Defini on The base a logarithm loga x is the inverse of the func on ax y = loga x ⇐⇒ x = ay The natural logarithm ln x is the inverse of ex . So y = ln x ⇐⇒ x = ey .
  • 87.
    Facts about Logarithms Facts (i) loga (x1 · x2 ) = loga x1 + loga x2
  • 88.
    Facts about Logarithms Facts (i) loga (x1 · x2 ) = loga x1 + loga x2 ( ) x1 (ii) loga = loga x1 − loga x2 x2
  • 89.
    Facts about Logarithms Facts (i) loga (x1 · x2 ) = loga x1 + loga x2 ( ) x1 (ii) loga = loga x1 − loga x2 x2 (iii) loga (xr ) = r loga x
  • 90.
    Logarithms convert productsto sums Suppose y1 = loga x1 and y2 = loga x2 Then x1 = ay1 and x2 = ay2 So x1 x2 = ay1 ay2 = ay1 +y2 Therefore loga (x1 · x2 ) = loga x1 + loga x2
  • 91.
    Examples Example Writeas a single logarithm: 2 ln 4 − ln 3.
  • 92.
    Examples Example Writeas a single logarithm: 2 ln 4 − ln 3. Solu on 42 2 ln 4 − ln 3 = ln 42 − ln 3 = ln 3 ln 42 not ! ln 3
  • 93.
    Examples Example 3 Write as a single logarithm: ln + 4 ln 2 4
  • 94.
    Examples Example 3 Write as a single logarithm: ln + 4 ln 2 4 Solu on 3 ln + 4 ln 2 = ln 3 − ln 4 + 4 ln 2 = ln 3 − 2 ln 2 + 4 ln 2 4 = ln 3 + 2 ln 2 = ln(3 · 22 ) = ln 12
  • 95.
    Graphs of logarithmicfunctions y y = 2x y = log2 x (0, 1) . (1, 0) x
  • 96.
    Graphs of logarithmicfunctions y y =y3= 2x x y = log2 x y = log3 x (0, 1) . (1, 0) x
  • 97.
    Graphs of logarithmicfunctions y y =y10y3= 2x =x x y = log2 x y = log3 x (0, 1) y = log10 x . (1, 0) x
  • 98.
    Graphs of logarithmicfunctions y y =x ex y =y10y3= 2x = x y = log2 x yy= log3 x = ln x (0, 1) y = log10 x . (1, 0) x
  • 99.
    Change of baseformula for logarithms Fact logb x If a > 0 and a ̸= 1, and the same for b, then loga x = logb a
  • 100.
    Change of baseformula for logarithms Fact logb x If a > 0 and a ̸= 1, and the same for b, then loga x = logb a Proof. If y = loga x, then x = ay So logb x = logb (ay ) = y logb a Therefore logb x y = loga x = logb a
  • 101.
    Example of changingbase Example Find log2 8 by using log10 only.
  • 102.
    Example of changingbase Example Find log2 8 by using log10 only. Solu on log10 8 0.90309 log2 8 = ≈ =3 log10 2 0.30103
  • 103.
    Example of changingbase Example Find log2 8 by using log10 only. Solu on log10 8 0.90309 log2 8 = ≈ =3 log10 2 0.30103 Surprised?
  • 104.
    Example of changingbase Example Find log2 8 by using log10 only. Solu on log10 8 0.90309 log2 8 = ≈ =3 log10 2 0.30103 Surprised? No, log2 8 = log2 23 = 3 directly.
  • 105.
    Upshot of changingbase The point of the change of base formula logb x 1 loga x = = · logb x = constant · logb x logb a logb a is that all the logarithmic func ons are mul ples of each other. So just pick one and call it your favorite. Engineers like the common logarithm log = log10 Computer scien sts like the binary logarithm lg = log2 Mathema cians like natural logarithm ln = loge Naturally, we will follow the mathema cians. Just don’t pronounce it “lawn.”
  • 106.
    “lawn” . . Image credit: Selva
  • 107.
    Summary Exponenals turn sums into products Logarithms turn products into sums Slide rule scabbards are wicked cool