Page 327 - Schaum's Outline of Theory and Problems of Advanced Calculus
P. 327

318                            IMPROPER INTEGRALS                         [CHAP. 12



                                                  ð  b               ð 1
                                               lim  f ðxÞ dx ¼1 and so  f ðxÞ dx diverges
                                               b!1 N                 a
                              For the cases where A ¼ 0 and A ¼1, see Problem 12.41.
                              As seen in this and the preceding problem, there is in general a marked similarity between proofs for
                          infinite series and improper integrals.

                                                       xdx               x   1
                                                 ð                   ð    2
                                                  1                   1
                     12.5. Test for convergence:  (a)          ;  ðbÞ  p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi dx.
                                                     4
                                                          2
                                                  1 3x þ 5x þ 1       2  x þ 16
                                                                          6
                                                                                 3
                                                                          4
                          (a) Method 1:  For large x,the integrand is approximately x=3x ¼ 1=3x .
                                          x         1     1  ð  1 dx
                                 Since          @    , and      converges ( p integral with p ¼ 3), it follows by the
                                        4
                                            2
                                      3x þ 5x þ 1  3x 3   3 1 x 3
                                             ð
                                              1    xdx
                              comparison test that        also converges.
                                                      2
                                                  4
                                              1 3x þ 5x þ 1
                                 Note that the purpose of examining the integrand for large x is to obtain a suitable comparison
                              integral.
                                                    x           1                 1     ð 1
                                                                 .  Since lim  f ðxÞ  ¼ , and  gðxÞ dx converges,
                              Method 2: Let f ðxÞ¼  4  2  ; gðxÞ¼  3
                                                3x þ 5x þ 1     x                 3
                              ð                                          x!1 gðxÞ       1
                               1
                                f ðxÞ dx also converges by the quotient test.
                               1
                                                                                     1
                                 Note that in the comparison function gðxÞ,we have discarded the factor .Itcould, however, just
                                                                                     3
                              as well have been included.
                                                  x       1
                              Method 3:  lim x 3         ¼ .  Hence, by Theorem 1, Page 309, the required integral
                                               4
                                                    2
                                             3x þ 5x þ 1  3
                                       x!1
                              converges.
                                                                           ffiffiffiffiffi
                                                                        2
                                                                         p
                                                                            6
                          (b) Method 1:  For large x,the integrand is approximately x = x ¼ 1=x.
                                            2
                                                                             2
                                           x   1   1 1      1  ð  1 dx   ð 1  x   1
                                           ffiffiffiffiffiffiffiffiffiffiffiffiffi A    .  Since  diverges,  ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi dx also diverges.
                                 For x A 2, p                              p
                                                                              6
                                            6
                                           x þ 1   2 x      2 2 x         2  x þ 16
                                                   2
                                                  x   1       1                          ð  1
                              Method 2:  Let f ðxÞ¼ p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi, gðxÞ¼ .  Then since lim  f ðxÞ  ¼ 1, and  gðxÞ dx diverges,
                                                   6
                                                  x   16      x           x!1 gðxÞ        2
                              ð
                               1
                                f ðxÞ dx also diverges.
                               2
                                                       !
                                                   2
                                                  x   1
                              Method 3:  Since lim x p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 1, the required integral diverges by Theorem 1, Page 309.
                                                  x þ 16
                                           x!1     6
                                 Note that Method 1 may (and often does) require one to obtain a suitable inequality factor (in this
                                 1
                                                           1
                              case ,or any positive constant less than )before the comparison test can be applied. Methods 2 and
                                 2                         2
                              3, however, do not require this.
                                    ð
                                    1    2
                     12.6. Prove that  e  x  dx converges.
                                    0
                                  2  x
                              lim x e  2  ¼ 0(by L’Hospital’s rule or otherwise). Then by Theorem 1, with A ¼ 0; p ¼ 2 the given
                              x!1
                          integral converges. Compare Problem 11.10(a), Chapter 11.
                     12.7. Examine for convergence:
                                 ln x                                     1   cos x
                              ð                                        ð
                               1                                        1
                                      dx; where a is a positive constant;         dx:
                               1 x þ a                                  0    x
                          ðaÞ                                       ðbÞ       2
                                    ln x
                          (a)  lim x    ¼1.   Hence by Theorem 1, Page 309, with A ¼1; p ¼ 1, the given integral diverges.
                                    x þ a
                              x!1
   322   323   324   325   326   327   328   329   330   331   332