P-SET FOR MA412 DUE 2/9/2012

JOE STAHL

1. 18.3 Let n be a positive integer and let P (z ) and Q(z ) be polynomials, where Q(z0 ) = 0. Use Theorem 2 in Sec. 16, as well as limits appearing in that section, to nd 1 (a) lim n (z0 = 0) z z0 z If z0 = 0, then z n is continuous and nonzero at z0 , and is nonzero in a neighborhood of z0 (by theorem 18.2). Then by theorem 16.2, taking f (z ) = 1 and F (z ) = z n , we have f (z0 ) = 1 and n limzz0 F (z0 ) = z0 = 0, so 1 1 lim n = n z z0 z z0 iz 3 1 z i z + i Referring to theorem 16.2 with f (z ) = iz 3 1 and F (z ) = z + i, we see that limzi f (z ) = i i3 1 = 0 and limzi F (z ) = i + i = 2i = 0, so (b) lim iz 3 1 0 = =0 z i z + i 2i lim P (z ) Q(z ) Appealing again to theorem 16.2 and taking f (z ) = P (z ) and F (z ) = Q(z ), we see limzz0 P (z ) = P (z0 ) and limzz0 Q(z ) = Q(z0 ) = 0, so (c) lim
z z0 z z0

lim

P (z ) P (z0 ) = Q(z ) Q(z0 )

2. 18.5 z 2 z as z tends to 0 does not exist. Do this by letting nonzero points z = (x, 0) and z = (x, x) approach the origin. Let P1 be the path that tends to 0 along the x axis, that is, the path dened by z = (x, 0), x R. Then the limit as z tends to 0 along P1 is f (z ) =
x 0

Show that the limit of the function

lim

x + i0 x i0

= lim
1

x 0

x x

= lim 1 = 1
x0

JOE STAHL

Let P2 be the path that tends to 0 along the line x = y (dened by z = (x, x), x R. Then the limit as z tends to 0 along P2 is lim x + ix x ix
2

x0

= lim

x 0

x(1 + i) x(1 i)

= lim

1+i 1+i = =1 x0 1 i 1i

Since the two limits are not equal, the limit in question does not exist.

3. 20.1 Use results in Sec. 20 to nd f (z ) when (a) f (z ) = 3z 2 2z + 4 f (z ) = 2 3z 1 1 2z 0 + 0 = 6z 2

(b) f (z ) = 1 4z 2 f (z ) = 3 1 4z 2
2

2 4 z 1
2

= 24z 1 4z 2

(c) f (z ) =

1 z1 (z = ) 2z + 1 2 (z = 1 ) 2

f (z ) = =

1 (2z + 1) (z 1) 2 (2z + 1) 3 (2z + 1)

2 2

(z =

1 ) 2

1 + z2 (d) f (z ) = z2 4 1 + z2 2 1 + z2
3 3

(z = 0)
4

f (z ) = =

(2 z ) z 2 2 z 1 + z 2 (z 2 ) 3z 2 1
2

(z = 0)

z3

(z = 0)

P-SET FOR MA412 DUE 2/9/2012

4. 20.3 Apply denition (3), sec. 19, of derivative to give a direct proof that 1 dw 1 = 2 when w = (z = 0) dz z z w = 1 1 z + z z dw w = lim z 0 dz z 1 = lim z+z z 0 z = lim
z z z z (z +z ) z 0

1 z

z z = lim z 0 z z (z + z ) 1 = lim z 0 z (z + z ) 1 = z (z + 0) 1 = 2 (z = 0) z

5. 20.4 Suppose that f (z0 ) = g (z0 ) = 0 and that f (z0 ) and g (z0 ) exist, where g (z0 ) = 0. Use denition (1), Sec. 19, of derivative to show that
z z0

lim

f (z ) f (z0 ) = g (z ) g (z0 ) Also note that if z = z0 , then

f (z )f (z0 ) z z0 g (z )g (z0 ) z z0

If f (z0 ) = g (z0 ) = 0, then

f (z ) g (z )

f (z )f (z0 ) g (z )g (z0 ) .

f (z ) f (z ) f (z0 ) f (z ) f (z0 ) z z0 = = = g (z ) g (z ) g (z0 ) g (z ) g (z0 ) z z0 Thus, lim f (z ) = lim g (z ) zz0 = =

f (z )f (z0 ) z z0 g (z )g (z0 ) z z0 )f (z0 ) limzz0 f (zz z0 )g (z0 ) limzz0 g(zz z0

z z0

f (z0 ) g (z0 )

Because by assumption f (z0 ) and g (z0 ) exist, this holds true and we are done.

JOE STAHL

6. 20.8 Use the method in Example 2, Sec. 19, to show that f (z ) does not exist at any point z when (a) f (z ) = Re z Note that Re z =
z +z 2 .

Then if f exists, f (z ) = lim = lim

z +z +z +z 2

z +z 2

z 0

z
z +z +z +z z z 2

z z + z = lim z 0 2z 1 z = lim + z 0 2 2z
z 0

However, the second part of this limit has already been shown not to exist (Sec. 19, Ex. 2, pg. 57), so the limit does not exist at any arbitrary point z , and thus the derivative fails to exist. (b) f (z ) = Im z First note that Im z =
z z 2i .

Then if f exists, f (z ) = lim = lim

z +z z +z 2i

z z 2i z z 2i

z 0

z
z +z z +z 2i

z z + z z + z z + z = lim z 0 2iz z + z = lim z 0 2iz 1 z = lim + z 0 2i 2iz

z 0

But again, the second part of this limit does not exist as shown in Sec. 19, Ex. 2, pg. 57, so the derivative fails to exist at any arbitrary point z C.

7. 20.9 Let f denote the function whose values are f (z ) =

z2 z

when z = 0 when z = 0

w Show that if z = 0, then z = 1 at each nonzero point on the real and imaginary axes in the z , w or xy , plane. Then show that z = 1 at each nonzero point (x, x) on the line y = x in that plane. Conclude from these observations that f (0) does not exist.

P-SET FOR MA412 DUE 2/9/2012

Letting z = x + iy tend to 0 along the x (x) axis, we nd df |z=(x,0) = lim x0 dz

(0+x)2 0+x

x = lim x0 x = lim 1
x0

=1 However, if z tends to 0 along the line y = x, then the limit is df dz = lim

z =(x,x) (xiy )2 x+ix

x0

x + ix

(x ix)2 x0 (x + ix)2 x2 (1 i)2 = lim x0 x2 (1 + i)2 = lim 1i x0 1 + i = lim 1 = lim

x0 2

= 1 But since 1 = 1, the derivative does not exist at z = 0.

8. 23.1 Use the theorem in Sec. 21 to show that f (z ) does not exist at any point if (a) f (z ) = z z = x iy ux = 1 vy = 1 uy = 0 vx = 0 ux = vy for any choice of z0 , so by theorem 21.1, neither does f . (b) f (z ) = z z z z = x + iy (x iy ) = 2iy ux = 0 vy = 2 uy = 0 vx = 0

JOE STAHL

ux = vy for any choice of z0 , so by theorem 21.1, neither does f . (c) f (z ) = 2x + ixy 2 f (z ) = 2x + ixy 2 ux = 2 vy = 2xy uy = 0 vx = y 2
1 Although we can equate ux = vy by letting z be on the unit hyperbola 1 = xy , that choice of z can never satisfy the requirement that vx = 0, so for any choice of z0 , so by theorem 21.1, f does not exist for any z C.

(d) f (z ) = ex eiy f (z ) = ex eiy = ex (cos y i sin y ) = ex cos y iex sin y ux = ex cos y vy = ex cos y uy = ex sin y vx = ex sin y By these equations, the only time that f could possibly exist is when cos y = 0 and simultaneously sin y = 0, because the only way ux = vy and uy = vx is if cos y = cos y and similarly for sin. However, cosine and sine have nonintersecting sets of zeros, and so ux = vy and uy = vx never happen at the same time. Thus, f fails to exist at any z C.

9. 23.2 Use the theorem in Sec. 21 to show that f (z ) and its derivative f (z ) exist everywhere, and nd f (z ) when (b) f (z ) = ex eiy f (z ) = ex eiy = ex (cos y i sin y ) = ex cos y iex sin y ux = ex cos y vy = ex cos y uy = ex sin y vx = ex sin y

P-SET FOR MA412 DUE 2/9/2012

Since ux = vy and uy = vx for all z C, f exists everywhere. f (z ) = ex cos y + iex sin y ux = ex cos y vy = ex cos y uy = ex sin y vx = ex sin y Where this time u and v represent the real and imaginary parts of f . One can see that the partials of f also satisfy the Caucy-Riemann equations at any point z , and so f exists everywhere as well. Using the partials found, we see that f (z ) = ex cos y iex sin y = f (z )

(d) f (z ) = cos x cosh y i sin x sinh y f (z ) = cos x cosh y i sin x sinh y ux = sin x cosh y vy = sin x cosh y uy = cos x sinh y vx = cos x sinh y The partials of f obviously satisfy the Cauchy-Riemann equations everywhere, and so f exists everywhere. Taking derivatives again, we see f (z ) = sin x cosh y i cos x sinh y ux = cos x cosh y vy = cos x cosh y uy = sin x sinh y vx = sin x sinh y The partials of the real and imaginary parts of f also satisfy the Cauchy-Riemann equations everywhere, so f exists everywhere and is equal to f (z ) = cos x cosh y + i sin x sinh y = f (z )

10. 23.3 From results obtained in Secs. 21 and 22, determine where f (z ) exists and nd its value when (b) f (z ) = x2 + iy 2 f (z ) = x2 + iy 2 ux = 2x vy = 2y uy = 0 vx = 0

JOE STAHL

Since f exists when the partials of f satisfy the Cauchy-Riemann equations, ux must equal vy at z0 if f exists at z0 . Then f (z ) exists on the line x = y , because that is when the equations are satised.

(c) f (z ) = z Im z

f (z ) = xy + iy 2 ux = y vy = 2y uy = x vx = 0 By the Cauchy-Riemann equations, f (z ) will exist when y = 2y (or equivalently, when y = 0) and when x = 0, so f (z ) exists on the coordinate axes.

11. 23.5 Show that when f (z ) = x + i (1 y ) , it is legitimate to write f (z ) = ux + ivx = 3x2 only when z = i. f (z ) = x3 + i (1 y ) ux = 3x2 vy = 3 (1 y ) uy = 0 vx = 0 If f (z ) exists, then 3x2 = 3 (1 y ) , so x2 = (1 y ) . Since we have a real square equal to the negative of another real square, they must both equal 0. Then it follows that x = 0 and y = 1, so f (z ) exists only when z = i.
2 2 2 3 3 3

12. 23.6 Let u and v denote the real and imaginary components of the function f dened by means of the equations f (z ) =
z2 z

when z = 0 when z = 0

P-SET FOR MA412 DUE 2/9/2012

Verify that the Cauchy-Riemann equations ux = vy and uy = vx are satised at the origin z = (0, 0). z2 (x iy )2 = z x + iy 2 x y 2 2ixy = x + iy 2 (x y 2 2ixy )(x iy ) = (x + iy )(x iy ) 3 x 3xy 2 + i(3x2 y + y 3 ) = x2 + y 2 z2 x3 3xy 2 = z x2 + y 2 z2 3x2 y + y 3 = z x2 + y 2
x3 3x(02 ) x2 +02

Re Im

u(0 + x, 0) u(0, 0) lim = lim x0 x0 x = lim =1 v (0, 0 + y ) v (0, 0) = lim lim y 0 y 0 y

x0

x x x
3(02 )y +y 3 02 +y 2

y = lim y 0 y =1 u(0, 0 + y ) u(0, 0) lim = lim y 0 y 0 y = lim =0 v (0 + x, 0) v (0, 0) lim = lim x0 x0 x

3x2 (0)+03 x2 +02 y 0 03 3(0)y 2 02 +y 2

y 0 y 0

0 = lim x0 x =0 From these equations, we can see that ux = vy and uy = vx at z = 0. 13. 23.7 Solve equations (2), Sec. 23 for ux and uy to show that ux = ur cos u sin r cos uy = ur sin + u r

10

JOE STAHL

Then use these equations and similar ones for vx and vy to show that in Sec. 23 equations (4) are satised at a point z0 if equations (6) are satised there. Thus complete the verication that equations (6), Sec. 23, are the Cauchy-Riemann equations in polar form. Solving the equations in Sec. 23, pg.69, we get ur cos = ux cos2 + uy sin cos u sin = ux r sin2 + uy r cos sin ur cos u sin = ux cos2 + uy sin cos + ux sin2 uy cos sin r = ux

ur sin = ux cos sin + uy sin2 u cos = ux r sin cos + uy r cos2 ur sin + u cos = ux cos sin + uy sin2 ux sin cos + uy cos2 r = uy

We also see vr cos = vx cos2 + vy sin cos v sin = vx r sin2 + vy r cos sin vr cos v sin = vx cos2 + vy sin cos + vx sin2 vy cos sin r = vx

vr sin = vx cos sin + vy sin2 v cos = vx r sin cos + vy r cos2 vr sin + v cos = vx cos sin + vy sin2 vx sin cos + vy cos2 r = vy

To show that if r ur = v and u = r vr are satised at a point z0 when the Cauchy-Riemann equations are satised at a point z0 , notice that r ur = ux r cos + uy r sin = vy r cos vx r sin = v u = ux r sin + uy r cos = r (ux sin uy cos ) = r (vy sin + vx cos ) = r vr