Mott scattering

[juniorpena]

This change is implementing Mott scattering of high energy electrons off of Helium nuclei. I uploaded the files from my Kassiopeia setup on my local machine into a repo I forked, and I made changes to the appropriate CMakeLists.txt files.

[zykure]

Thanks for the PR and for providing documentation with the code! I have some comments that need to be addressed before merging.

[zykure]

Is there any update on this PR @juniorpena ?

[juniorpena]

Hi, No updates as of now. Other projects came to the forefront, and I still have not gone back to update this code. I will probably not get back to this until sometime closer to the summer. Best, Junior From: J B ***@***.***> Reply-To: KATRIN-Experiment/Kassiopeia ***@***.***> Date: Monday, January 9, 2023 at 3:29 AM To: KATRIN-Experiment/Kassiopeia ***@***.***> Cc: Junior Ivan Pena ***@***.***>, Mention ***@***.***> Subject: Re: [KATRIN-Experiment/Kassiopeia] Mott scattering (PR #66) Is there any update on this PR @juniorpena<https://github.com/juniorpena> ? — Reply to this email directly, view it on GitHub<#66 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ARHKF652ECFCVFXCX6NP3TDWRPD6FANCNFSM57JDPJXQ>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[juniorpena]

Hello, I've eliminated the use of ROOT in this scattering calculator. Now the GetTheta function uses inverse transform sampling and rejection sampling. I've also updated it to allow for both He and Ne Mott Scattering

[2xB]

Hey, thank you very much! Could you add documentation of the added objects to https://github.com/KATRIN-Experiment/Kassiopeia/blob/main/Kassiopeia/XML/Complete/Interactions.xml ? That is helpful both for users and for the review.

(Related, but independent issue: #106 )

[juniorpena]

I've added the documentation

[2xB]

After looking through the code, I had three remaining comments after which this looks good to merge!

Btw. if there will be a thesis describing a use case of this, one can also advertise it in the top comment of KSIntCalculatorMott.h, since it is always useful to see what the original use case for components of Kassiopeia was.

[2xB]

For a fixed-size array, std::array is better than std::vector. However, in this case, I think the array is not needed, as you just build it and directly afterwards use it - one could just calculate theta on the fly in the second loop. Also, it makes sense to document in a comment where the 1000 comes from and when it is valid?

[juniorpena]

It was arbitrarily chosen, but now there is a variable to control the size of the array and I added a comment about how it affects the simulation time

[2xB]

Thanks! I still believe one could avoid arrays/vectors completely along the lines of

#include <limits>

[...]

    // Find largest rescaling factor for rejection sampling over all angles
    double k = -std::numeric_limits<double>::infinity();

    for (int i = 0; i <= int(resolution); ++i) {
        const double current_theta = fThetaMin + i * (fThetaMax - fThetaMin) / resolution;
        const double current_k     = MDCS(theta, anEnergy) / Normalized_RDCS(theta);

        if (k < current_k) {
            k = current_k;
        }
    }

, but I'm not sure if that is relevant performance- or memory-wise at all.

[2xB]

Our Ubuntu 20.04 CI check failed with

/__w/Kassiopeia/Kassiopeia/Kassiopeia/Interactions/Source/KSIntCalculatorMott.cxx:10:7: error: use of undeclared identifier 'KGeoBag'
using KGeoBag::KThreeVector;
      ^
/__w/Kassiopeia/Kassiopeia/Kassiopeia/Interactions/Source/KSIntCalculatorMott.cxx:43:5: error: unknown type name 'KThreeVector'; did you mean 'katrin::KThreeVector'?
    KThreeVector tPosition = anInitialParticle.GetPosition();
    ^~~~~~~~~~~~
    katrin::KThreeVector

, as can be seen by clicking on the check details...

[2xB]

Btw. additionally to fixing the above error, I believe that it would make sense to update your branch by merging our main branch into it, that would fix the fedora_37 test case.

[2xB]

This is for ²⁰Ne and the above for ⁴He, right? Is your entire code specific to those isotopes?

[2xB]

@juniorpena What is the status here? I have added one question to resolve one of the two open threads here, the other I think is not that important but I leave it open just in case you want to comment on it as well. I think once at least the thread regarding the provided isotopes is solved, we should finally be able to merge this from my side. (CC @richeldichel )