Scribd
Upload
29 views2 pages

Eigono Da

This document contains summaries of several appendices related to deriving and computing equations for diffusion problems. It discusses: 1) Using asymptotic expressions for Bessel functions to evaluate a residue. 2) Breaking integrals into regions and simplifying functions using identities to derive Equation 27. 3) Computing integrals as alternating series based on zeros of integrands and using Gaussian quadrature. 4) Deriving the solution for constant diffusivity by taking the Laplace transform and choosing b so the solution is finite as r approaches infinity. 5) Rewriting Equation 27 and taking limits as the product of r and the diffusion coefficient approach zero to derive the analytical solution for that case.

Uploaded by

Romy Szorga
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
29 views2 pages

Eigono Da

This document contains summaries of several appendices related to deriving and computing equations for diffusion problems. It discusses: 1) Using asymptotic expressions for Bessel functions to evaluate a residue. 2) Breaking integrals into regions and simplifying functions using identities to derive Equation 27. 3) Computing integrals as alternating series based on zeros of integrands and using Gaussian quadrature. 4) Deriving the solution for constant diffusivity by taking the Laplace transform and choosing b so the solution is finite as r approaches infinity. 5) Rewriting Equation 27 and taking limits as the product of r and the diffusion coefficient approach zero to derive the analytical solution for that case.

Uploaded by

Romy Szorga
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Appendix C-Computation of Residue at s=0

Res
Exp
Also
Consequently, using the asymptotic expressions for Bessel function of large arguments 5, we have
Appendix D-Derivation of Equation 27
The following identities are used 7:
Because 0<p2, we must break the integral into the sum of three integrals with integration limits defined
as follows.
Region I.
If r = 0, the contribution from region III vanishes and only two integrals remain.
In these regions, the arguments of the Bessel functions takes either the from F exp(+j) or the from F
exp(+k/2).
We can simplify the functions by using identities in eqs. D-1 through D-4.
Recall that
Along the branch EF, s=p exp ( r ), and Eqs. D-5 through D-7 can be simplified as follows.
Combining Eqs. D-1 through D-25, we have the following.
Now we are ready to evaluate Eq. B-6. Using Eq. B-2,
And substituting s=p e , we obtain.
The integration interval of Eq. D-35 is broken into three parts, corresponding to the three regions, III, II,
and I, respectively.
Now we can simplify ^ in each region.
Note that a,b,c,d,e, and f are all complex. Similarly , L1 can be simplified by substituting s=pe. following
the previous procedure, we would find that L1 = -L2 , where L2 is the complex conjugate of L 2 .
therefore , the final result, Eq. 27, is

Appendix E – Computing the Integrals


The oscillatory nature of the integrands 0f M 1,M2,M3,N1,and N2 suggests thet these integrals
should be com-puted as an alternating series. Let w0=0 and Wn be the value of w corresponding
to the nth zero of G (w), where G is the integrand of integrals. Then the integral in the analytical
solution can be written as
Where h is the magnitude of the integral over the half cycle from w, the zeros of g can be
determined by any Newtonian-interation method, the integrals over the half cycle can be
evaluated by a Gaussian quadrature method with either 20 or 64 gauss point.
Appendix F-Derivation of Solution for Constant D
In the case of constant diffusivity , Eq. 4 reduces to
The laplace transform of Eq. F-1 with the initial con-centration equal to zero is

That Solution of Eq. F-2 which is finite as r -00 is

Where K a is the modifiend Bessel function of order a . We want to choose b so thet l (c) = 1/s
as r-0.

The inverse of l (c) can be found from carslaw and jaeger8:

Appendix G-Derivation of the Analitical Solution With rw=0


First we can rewrite Eq.27 as

Consider the first term in the brackats:

If rw =rDw=0,Eq.G-2 can be rewritte as


similary

As rDw-0,the above equation becomes

You might also like