Cut 2D-histogram

Newbie
, ,
1

Hi,
I am currently trying to analyze the energy loss of electrons in an active target gas. When I drew a 2D histogram with the ∆E/∆s versus ∑∆s (i.e. energy loss per step divided by the length of the step, versus the accumulated depth of the electron) I get the following histogram:

2d_histo.pdf (303.4 KB)

I don’t understand where this accumulation of points is coming from (it seems pretty unrealistic). Besides, the histogram seems to end sharply just above it. I do not know what is causing these effects

ROOT Version: 6.22/07
Platform: MacOS
Compiler: Not Provided

2

By “accumulation of points” I meant the thick horizontal bar located around 60 keV/mm

3

Hi Javier,

true, both effects (the excess at 60 keV/mm and the cut-off at 75 keV/mm) look unrealistic. I’d suggest to check the code that produces your histogram.

1 Like
4

Hi yus, here is a piece of the code:

    string str;
    while(getline(*file, str))

    {
        *file >> stepNumber >> Edep >> stepLength;

        stepNumbervec.push_back(stepNumber);
        Edepvec.push_back(Edep);
        stepLengthvec.push_back(stepLength);
    }



    TCanvas *c2 = new TCanvas("c2", "Bragg canvas");
    Int_t bragg_bins = 1000; // bins of histogram
    TH2F *bragg_hist = new TH2F("bragg_hist", Form("%s, %.2f MeV, %.0f K, %.0f mbar, %.0f V/cm", string(gas).c_str(), initial_energy, temperature, pressure, electric_field), bragg_bins, 0, 1400, bragg_bins, 0, 200);
    
    
    Double_t depth = 0; // accumulated depth
    Double_t stoppingPower;
    
    for(Int_t i=0; i<Edepvec.size(); i++)
    {
        if (stepNumbervec[i] != 1)
        {
            // stick to the same electron? -> add the step length to the total range
            depth += stepLengthvec[i];
            stoppingPower = Edepvec[i]*1000/stepLengthvec[i];
            c2->cd();
            bragg_hist->Fill(depth, stoppingPower);
        }
        else
        {
            // another electron? -> filling the last step of the previous one and clear depth
            c1->cd();
            range_hist->Fill(depth);
            depth = 0;
        }
    }
    
    c2->cd();
    //bragg_hist->Rebin2D(4, 4);
    bragg_hist->Draw("COLZ");

I didn’t upload it originally because it seemed pretty straightforward to me… The code reads a .txt file with the step number of the electron trajectory, the energy lost in that step and the length of the step, for example:
1 0.000125801 0.325053
2 0.0325084 74.1814
3 0.052952 120.936
4 0.0850733 226.414
5 0.0120996 31.1658
(where the second column is in MeV, and the third is in mm).
If you spot any problem please let me know. Otherwise I guess I will have to start looking at the Geant4 simulation that generates the file.

Thanks in advance

5

How large is this text file of yours? Can you post it here or maybe share it via your favorite file-sharing service like Dropbox/Google Drive/Mega etc.?

6

It is quite large actually. This is the link to the file uploaded to google drive:

I must say, however, that when I ran the simulation for protons, just to see if I recovered the Bragg peak, I got this histogram for 3 MeV protons:

proton
proton698×447 65.7 KB

And none of the effects I observed with the electrons histogram appeared.

7

Yes, I reproduced the plot. There is probably something wrong with the way your text file was generated.

1 Like
8

Yes, that is what I am looking at right now. Guess I’ll have to post a topic in a Geant4 forum :laughing:. Thank you anyway for the help!