7 Error Functions, Dawson’s and Fresnel IntegralsProperties

§7.13 Zeros

Contents
  1. §7.13(i) Zeros of erf⁡z
  2. §7.13(ii) Zeros of erfc⁡z
  3. §7.13(iii) Zeros of the Fresnel Integrals
  4. §7.13(iv) Zeros of ℱ⁡(z)

§7.13(i) Zeros of erf⁡z

erf⁡z has a simple zero at z=0, and in the first quadrant of ℂ there is an infinite set of zeros zn=xn+i⁢yn, n=1,2,3,…, arranged in order of increasing absolute value. The other zeros of erf⁡z are −zn, z¯n, −z¯n.

Table 7.13.1 gives 10D values of the first five xn and yn. For graphical illustration see Figure 7.3.5.

Table 7.13.1: Zeros xn+i⁢yn of erf⁡z.
n xn yn
1 1.45061 61632 1.88094 30002
2 2.24465 92738 2.61657 51407
3 2.83974 10469 3.17562 80996
4 3.33546 07354 3.64617 43764
5 3.76900 55670 4.06069 72339

As n→∞

7.13.1 xn ∼λ−14⁢μ⁢λ−1+116⁢(1−μ+12⁢μ2)⁢λ−3−⋯,
yn ∼λ+14⁢μ⁢λ−1+116⁢(1−μ+12⁢μ2)⁢λ−3+⋯,

where

7.13.2 λ =(n−18)⁢π,
μ =ln⁡(λ⁢2⁢π).

§7.13(ii) Zeros of erfc⁡z

In the sector 12⁢π<ph⁡z<34⁢π, erfc⁡z has an infinite set of zeros zn=xn+i⁢yn, n=1,2,3,…, arranged in order of increasing absolute value. The other zeros of erfc⁡z are z¯n. The zeros of w⁡(z) are i⁢zn and i⁢z¯n.

Table 7.13.2 gives 10D values of the first five xn and yn. For graphical illustration see Figure 7.3.6.

Table 7.13.2: Zeros xn+i⁢yn of erfc⁡z.
n xn yn
1 −1.35481 01281 1.99146 68428
2 −2.17704 49061 2.69114 90243
3 −2.78438 76132 3.23533 08684
4 −3.28741 07894 3.69730 97025
5 −3.72594 87194 4.10610 72847

As n→∞

7.13.3 xn ∼−λ+14⁢μ⁢λ−1−116⁢(1−μ+12⁢μ2)⁢λ−3+⋯,
yn ∼λ+14⁢μ⁢λ−1+116⁢(1−μ+12⁢μ2)⁢λ−3+⋯,

where

7.13.4 λ =(n−18)⁢π,
μ =ln⁡(2⁢λ⁢2⁢π).

§7.13(iii) Zeros of the Fresnel Integrals

At z=0, C⁡(z) has a simple zero and S⁡(z) has a triple zero. In the first quadrant of ℂ C⁡(z) has an infinite set of zeros zn=xn+i⁢yn, n=1,2,3,…, arranged in order of increasing absolute value. Similarly for S⁡(z). Let zn be a zero of one of the Fresnel integrals. Then −zn, z¯n, −z¯n, i⁢zn, −i⁢zn, i⁢z¯n, −i⁢z¯n are also zeros of the same integral.

Tables 7.13.3 and 7.13.4 give 10D values of the first five xn and yn of C⁡(z) and S⁡(z), respectively.

Table 7.13.3: Complex zeros xn+i⁢yn of C⁡(z).
n xn yn
1 1.74366 74862 0.30573 50636
2 2.65145 95973 0.25290 39555
3 3.32035 93363 0.22395 34581
4 3.87573 44884 0.20474 74706
5 4.36106 35170 0.19066 97324

As n→∞ the xn and yn corresponding to the zeros of C⁡(z) satisfy

7.13.5 xn ∼λ+α⁢(α⁢π−4)8⁢π⁢λ3+⋯,
yn ∼α2⁢λ+⋯,

with

7.13.6 λ =4⁢n−1,
α =(2/π)⁢ln⁡(π⁢λ).
Table 7.13.4: Complex zeros xn+i⁢yn of S⁡(z).
n xn yn
1 2.00925 70118 0.28854 78973
2 2.83347 72325 0.24428 52408
3 3.46753 30835 0.21849 26805
4 4.00257 82433 0.20085 10251
5 4.47418 92952 0.18768 85891

As n→∞ the xn and yn corresponding to the zeros of S⁡(z) satisfy (7.13.5) with

7.13.7 λ =2⁢n,
α =(2/π)⁢ln⁡(π⁢λ).

§7.13(iv) Zeros of ℱ⁡(z)

In consequence of (7.5.5) and (7.5.10), zeros of ℱ⁡(z) are related to zeros of erfc⁡z. Thus if zn is a zero of erfc⁡z (§7.13(ii)), then (1+i)⁢zn/π is a zero of ℱ⁡(z).

For an asymptotic expansion of the zeros of ∫0zexp⁡(12⁢π⁢i⁢t2)⁢dt (=ℱ⁡(0)−ℱ⁡(z) =C⁡(z)+i⁢S⁡(z)) see Tuẑilin (1971).