Solving Asymptotic Formula: Eq. 25 & 27

In summary, the conversation discusses an equation for P(x;a) and how it can be simplified under certain conditions. By using equations 19, 20, and 24, the equation for P(x;a) is simplified to include terms involving Fresnel functions. With the additional assumption that N_F(a) is much smaller than 1 and that (x-aη)/aη is much greater than 1 over the square root of N_F(a)η, the equation can be further simplified to the final form of P(x;a) ≃ 2γ/π^2η^2 [a^2/(x^2/η^2-a^2)^2 + 1/(x^2/η^
  • #1
JBD
15
1
In the following equation,

$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{1}{π^2N_F(a)\eta(1 - \frac{x}{a\eta})^2} + \frac{1}{π^2N_F(a)\eta(1 + \frac{x}{a\eta})^2} +\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [sin (\frac{π N_F(a)\eta(1 - \frac{x}{a\eta})^2}{2})sin (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2}) + cos (\frac{πN_F(a)\eta(1 - \frac{x}{a\eta})^2}{2}
)cos (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2})]]$$if $$ 0<N_F(a) ≪ 1$$ and $$(x − aη)/aη ≫ 1/\sqrt{N_F (a)η}$$, how do you arrive at
$$P(x; a) ≃ \frac{2γ}{π^2η^2} (\frac{a^2}{(\frac {x^2}{η^2} − a^2)^2}
+\frac{1}{\frac{x^2}{η^2} − a^2} sin^2
(πN_F(a)\frac{x}{a}))$$Please see what I have done so far and check if I have errors in it.

Relevant Equations

$$\alpha (x; a) = \sqrt{N_F(a)\eta } (1 - \frac{x}{a\eta})$$Eq. 19

where $$\eta = 1 + L/D$$, $$N_F(a) = \frac{2a^2}{\lambda L}$$ and (additional definition) $$\gamma = \eta - 1$$

Starting with this:
Eq. 20
$$P(x; a)=\frac{1}{2\lambda(L+D)} ([C(α(x; a)) + C(α(x; −a))]^2 + [S(α(x; a)) + S(α(x; −a))]^2)$$
Eq. 24 (these two)
$$C[α(x; +a)] + C[α(x; −a)] ≃ \frac{1}{πα(x; a)} sin (\frac{πα(x; a)^2}{2}
) + \frac{1}{πα(x; -a)} sin (\frac{πα(x; -a)^2}{2}
) $$
and
$$S[α(x; +a)] + S[α(x; −a)] ≃ \frac{-1}{πα(x; a)} cos (\frac{πα(x; a)^2}{2}
) - \frac{1}{πα(x; -a)} cos (\frac{πα(x; -a)^2}{2}
) $$Here is what I have done so far:
$$(C[α(x; +a)] + C[α(x; −a)])^2 + (S[α(x; +a)] + S[α(x; −a)])^2 = \frac{1}{π^2α^2(x; a)} + \frac{1}{π^2α^2(x; -a)} + \frac{2}{π^2α(x; +a)α(x; −a)} [sin (\frac{πα(x; a)^2}{2})sin (\frac{πα(x; -a)^2}{2}) + cos (\frac{πα(x; a)^2}{2}
)cos (\frac{πα(x; -a)^2}{2})]$$

Using equation 19:

$$=\frac{1}{π^2N_F(a)\eta(1 - \frac{x}{a\eta})^2} + \frac{1}{π^2N_F(a)\eta(1 + \frac{x}{a\eta})^2} +\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [sin (\frac{π N_F(a)\eta(1 - \frac{x}{a\eta})^2}{2})sin (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2}) + cos (\frac{πN_F(a)\eta(1 - \frac{x}{a\eta})^2}{2}
)cos (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2})]$$

And then in equation 20, the outer factor:
$$\frac{1}{2\lambda (L+D)} = \frac{\gamma}{2\lambda L \eta}$$

So the new equation for P is:
$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{1}{π^2N_F(a)\eta(1 - \frac{x}{a\eta})^2} + \frac{1}{π^2N_F(a)\eta(1 + \frac{x}{a\eta})^2} +\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [sin (\frac{π N_F(a)\eta(1 - \frac{x}{a\eta})^2}{2})sin (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2}) + cos (\frac{πN_F(a)\eta(1 - \frac{x}{a\eta})^2}{2}
)cos (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2})]]$$But this is where I am not sure what to do anymore with $$N_F (a) ≪ 1$$ and if $$(x − aη)/aη ≫ 1/\sqrt{N_F (a)η}$$ to arrive at equation 25.I'm trying to verify equations 25 and 27 in the paper linked below. I got confused on how to apply $$N_F(a) ≪ 1$$ to get the final result. I was able to do "Applying the Fresnel function asymptotic forms (24) to (20) and using the definition (19)" but then I got stuck here

"we deduce that if $$N_F (a) ≪ 1$$ and if $$(x − aη)/aη ≫ 1/\sqrt{N_F (a)η}$$ , we
get the following asymptotic formula:"

Eq. 25
$$P(x; a) ≃ \frac{2γ}{π^2η^2} (\frac{a^2}{(\frac {x^2}{η^2} − a^2)^2}
+\frac{1}{\frac{x^2}{η^2} − a^2} sin^2
(πN_F(a)\frac{x}{a}))$$

I could not reproduce this result (eq 25 in the paper) as well as eq 27.Page 14 Equations 25 and 27
Here is the link https://arxiv.org/pdf/1110.2346.pdf
There are occasional typographical errors in the paper. (from what I have verified so far, pages 1-13) [1]: https://i.stack.imgur.com/Qc5Ie.png
 
Physics news on Phys.org
  • #2
JBD said:
But this is where I am not sure what to do anymore with NF(a)≪1
Hi JBD:

The only thing that occurs to me is to (1) replace sin(u) with u since u<<1, and (2) replace cos(u) with 1, since cos(u) ~= 1-u2.

Hope this helps.

Regards,
Buzz
 
  • Like
Likes JBD
  • #3
Buzz Bloom said:
Hi JBD:

The only thing that occurs to me is to (1) replace sin(u) with u since u<<1, and (2) replace cos(u) with 1, since cos(u) ~= 1-u2.

Hope this helps.

Regards,
Buzz
Thanks for helping. I did it differently and arrived at the final form but I had some assumptions that I don't know if allowed or not. Like for example, $$N_F(a) =\frac{2a^2}{\lambda L}$$ but since $$N_F(a) ≪1$$, I replaced it with $$\frac{a^2}{\lambda L}$$. Is this reasonable? If not, I would have arrived at a similar form but short of the factor 2. I also had to do it one more time in a fraction with large number in denominator. The numerator was one and I changed it to 2.
 
  • #4
JBD said:
In the following equation,

$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{1}{π^2N_F(a)\eta(1 - \frac{x}{a\eta})^2} + \frac{1}{π^2N_F(a)\eta(1 + \frac{x}{a\eta})^2} +\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [sin (\frac{π N_F(a)\eta(1 - \frac{x}{a\eta})^2}{2})sin (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2}) + cos (\frac{πN_F(a)\eta(1 - \frac{x}{a\eta})^2}{2}
)cos (\frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2})]]$$if $$ 0<N_F(a) ≪ 1$$ and $$(x − aη)/aη ≫ 1/\sqrt{N_F (a)η}$$, how do you arrive at
$$P(x; a) ≃ \frac{2γ}{π^2η^2} (\frac{a^2}{(\frac {x^2}{η^2} − a^2)^2}
+\frac{1}{\frac{x^2}{η^2} − a^2} sin^2
(πN_F(a)\frac{x}{a}))$$

##P(x;a)## has two parameters or arguments. When you go to the asymptotic form you are letting one of the arguments get large (or maybe small); would that be large (small?) ##x## or large (small?) ##a##?
 
Last edited:
  • Like
Likes JBD
  • #5
JBD said:
but since
NF(a)≪1​
I replaced it with
a2λL.​
Is this reasonable?

I underlined "it" in the quote because I am not sure what the antecedent of this pronoun is. I am guessing you intend it to be what is equal to NF(a). I also do not understand what would make this assumptions reasonable.

Could you post the details of your work that leads to the correct answer except for a factor of 2?

Regards,
Buzz
 
  • Like
Likes JBD
  • #6
Ray Vickson said:
##P(x;a)## has two parameters or arguments. When you go to the asymptotic form you are letting one of the arguments get large (or maybe small); would that be large (small?) ##x## or large (small?) ##a##?
$$x ≫ a$$ large x and small a
Sorry for the late reply.
 
Last edited:
  • #7
Buzz Bloom said:
I underlined "it" in the quote because I am not sure what the antecedent of this pronoun is. I am guessing you intend it to be what is equal to NF(a). I also do not understand what would make this assumptions reasonable.

Could you post the details of your work that leads to the correct answer except for a factor of 2?

Regards,
Buzz

I used cos (A-B) = cos A cos B + sin A sin B (I feel bad for not noticing this. I was too focused on applying the N_f(a) ≪1 and the other condition.)

$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{1}{π^2N_F(a)\eta(1 - \frac{x}{a\eta})^2} + \frac{1}{π^2N_F(a)\eta(1 + \frac{x}{a\eta})^2} +\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [cos (\frac{πN_F(a)\eta(1 - \frac{x}{a\eta})^2}{2}
- \frac{πN_F(a)\eta(1 + \frac{x}{a\eta})^2}{2})]]$$$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{2(1+\frac{x^2}{a^2\eta^2})}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})^2} +\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [cos (2πN_F(a)\frac{x}{a})]]$$$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{2}{π^2N_F(a)\eta(1 - \frac{x^2}{a^2\eta^2})} [\frac{1 + \frac{x^2}{a^2\eta^2}}{1 - \frac{x^2}{a^2\eta^2}}+cos (2πN_F(a)\frac{x}{a})]] $$$$P(x; a)= \frac{\gamma}{2\lambda L \eta} [\frac{2}{π^2N_F(a)\eta( \frac{x^2}{a^2\eta^2}-1)} [\frac{ \frac{x^2}{a^2\eta^2}}{ \frac{x^2}{a^2\eta^2}-1}+\frac{1 }{\frac{x^2}{a^2\eta^2}-1}-cos (2πN_F(a)\frac{x}{a})]] $$

$$\frac{ \frac{x^2}{a^2\eta^2}}{ \frac{x^2}{a^2\eta^2}-1} ≃ 1$$since$$ \frac{x}{a\eta}$$ is a large number$$P(x; a)≃ \frac{\gamma}{2\lambda L \eta} [\frac{2}{π^2N_F(a)\eta( \frac{x^2}{a^2\eta^2}-1)} [\frac{1 }{\frac{x^2}{a^2\eta^2}-1} +1-cos (2πN_F(a)\frac{x}{a})]] $$

I used 2 sin^2 A = 1 - cos 2A

$$P(x; a)≃ \frac{\gamma}{2\lambda L \eta} [\frac{2}{π^2N_F(a)\eta( \frac{x^2}{a^2\eta^2}-1)} [\frac{1 }{\frac{x^2}{a^2\eta^2}-1} +2 sin^2 (πN_F(a)\frac{x}{a})]] $$

$$N_F(a) = \frac{2a^2}{\lambda L}$$
so,
$$P(x; a)≃ \frac{\gamma}{ \eta} \frac{1}{π^22a^2\eta( \frac{x^2}{a^2\eta^2}-1)} [\frac{1 }{\frac{x^2}{a^2\eta^2}-1} +2 sin^2 (πN_F(a)\frac{x}{a})] $$
$$P(x; a)≃ \frac{\gamma}{π^2 \eta^2} (\frac{1}{2a^2( \frac{x^2}{a^2\eta^2}-1)} [\frac{1 }{\frac{x^2}{a^2\eta^2}-1} + 2sin^2 (πN_F(a)\frac{x}{a})]) $$
$$P(x; a)≃ \frac{\gamma}{π^2 \eta^2} (\frac{1}{2( \frac{x^2}{\eta^2}-a^2)} [\frac{a^2 }{\frac{x^2}{\eta^2}-a^2} + 2sin^2 (πN_F(a)\frac{x}{a})]) $$
$$P(x; a)≃ \frac{\gamma}{π^2 \eta^2} ( \frac{a^2 }{2(\frac{x^2}{\eta^2}-a^2)^2} + \frac{1}{( \frac{x^2}{\eta^2}-a^2) }sin^2 (πN_F(a)\frac{x}{a})) $$

A factor of 2 missing in front and an extra 2 in the denominator.
 
  • #8
Hi JDB:

I now see how you derived the answer you got, and where it is different from Eq 25. What still puzzles me is where the assumption
NF(a) = 2a2/λL​
comes from.

BTW, I now also see why my original suggestion is not helpful,

Regards,
Buzz
 
  • Like
Likes JBD
  • #9
Buzz Bloom said:
Hi JDB:

I now see how you derived the answer you got, and where it is different from Eq 25. What still puzzles me is where the assumption
NF(a) = 2a2/λL​
comes from.

BTW, I now also see why my original suggestion is not helpful,

Regards,
Buzz

Fresnel Number is defined as $$N_f(a) = a^2 / \lambda L$$
where a is the size of slit, lambda is wavelength and L is distance from slit to screen
In the paper, the slit was size 2a and distance from slit to screen is L + D, but D = L so
$$N_f(a) = 4a^2 / \lambda 2L = 2a^2/ \lambda L$$

I don't know if eq 25 has typos (because I encountered some of them ) but I think it does not have typos. But then again, where did I get my math wrong? Thanks.
 
  • #10
Hi JDB:

My guess is that the text you were using has typos or just plain errors. If the problem was part of some work project, I think you should just accept that you got the right answer. If the problem was related to study for a course, I suggest you review your work with the professor - he may feel inclined to give you some extra credit for finding an error in the text.

Regards,
Buzz
 

Related to Solving Asymptotic Formula: Eq. 25 & 27

1. What is an asymptotic formula?

An asymptotic formula is a mathematical expression that describes the behavior of a function as its input approaches a certain value. It is often used to estimate the growth rate or behavior of a function that may be too complex to calculate directly.

2. How do you solve an asymptotic formula?

To solve an asymptotic formula, you first need to identify the dominant term(s) in the expression. Then, you can use various techniques such as L'Hopital's rule, logarithmic differentiation, or simplification to reduce the expression to a more manageable form. Finally, you can take the limit of the simplified expression to find the asymptotic behavior.

3. What is Eq. 25 & 27 in the context of asymptotic formulas?

Eq. 25 and 27 refer to specific equations or expressions related to solving asymptotic formulas. These equations may vary depending on the specific problem or context, but they are often used as a shorthand notation to refer to a particular step in the solving process.

4. Can asymptotic formulas be used in real-world applications?

Yes, asymptotic formulas are commonly used in various fields such as physics, engineering, and computer science to analyze and predict the behavior of complex systems or functions. They can be used to estimate the performance of algorithms, model physical processes, and make predictions about the growth of populations or economies.

5. Are there limitations to using asymptotic formulas?

Yes, there are some limitations to using asymptotic formulas. They may not always provide an accurate representation of the exact behavior of a function, as they are based on approximations and simplifications. Additionally, they may not be applicable to all types of functions or systems, and their accuracy may decrease as the input values get larger or more complex.

Similar threads

Replies
4
Views
514
  • Differential Equations
Replies
5
Views
2K
  • Calculus and Beyond Homework Help
Replies
7
Views
730
  • Calculus and Beyond Homework Help
Replies
9
Views
1K
  • Special and General Relativity
Replies
1
Views
978
  • Calculus and Beyond Homework Help
Replies
13
Views
3K
  • Calculus and Beyond Homework Help
Replies
8
Views
2K
  • Calculus and Beyond Homework Help
Replies
3
Views
1K
  • Calculus and Beyond Homework Help
Replies
4
Views
1K
Replies
1
Views
164
Back
Top