numerical.py 4.84 KB
Newer Older
Matheus Müller's avatar
Matheus Müller committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
#-*- coding: utf-8 -*-
# Comparação do espectros de espalhamento baseados na função modificada de Bessel do primeiro tipo de ordem 1 
# com o modelo Gaussiano. 
# Referência: R. P. Pezzi, et al. Applied Physics Letters, 92 164102 (2008)

from pylab import *
from numpy import nan_to_num
from scipy.special import i1
from Ewindow import Edict

PI=math.pi

####################################################################################################################################################
# Seção de choque Rutherford
def CSRf(Z1, Z2, Ein, M1, M2, Theta):
    if M1 < M2:
        return ((Z1*Z2*4.8e-20*1e8)/(4*Ein))**2 * sin(Theta)**-4 * (sqrt(1-((M1/M2)*sin(Theta))**2)+cos(Theta))**2 / sqrt(1-((M1/M2)*sin(Theta))**2) 
    else:
        return 0

####################################################################################################################################################
# Função de convolução com forma de linha
def convoluiF0(espectro, alpha,passoE):

    areaantes = espectro.sum()

    temp = espectro[::-1]
    eixo = arange(0,12.*alpha,passoE)
    lineshape = alpha*exp(-eixo/alpha)

#    ls_plt=plt.plot(lineshape/lineshape.sum(),lw=2)
#    sig_plt=plt.plot(temp/temp.sum(),lw=2)

    temp = convolve(temp/temp.sum(), lineshape/lineshape.sum())
    temp.resize(len(espectro))

#    result_plt=plt.plot(temp/temp.sum(),'o')
#    legend([ls_plt,sig_plt,result_plt],["Lineshape","Sinal","Resultado"])

    return temp[::-1]*areaantes

####################################################################################################################################################
def spectro(p, modelo, ion, Emin, Emax, EPasso, passox, LScontrol):

    e = arange(Emin, Emax, EPasso)
    Y = e*0.

    Theta_s = PI - (p['Theta_in'] + p['Theta_out'])*PI/180
    mi = ion['mass']
    espessura = 15

    ## Calcula o espectro de cada elemento da amostra
    for componente in Edict:
   
        mt = Edict[componente]['mass']
        c = Edict[componente]['dist']
        alpha_lineshape = Edict[componente]['LineShape']

        k = ((sqrt (mt**2+mi**2*(sin(Theta_s)**2)) + mi*cos(Theta_s) ) / (mi+mt) )**2	
        dedxef=p['dedx']*(k/cos(p['Theta_in']*PI/180.) + 1/cos(p['Theta_out']*PI/180.))
        dw2dxef=p['dW2dx']*(k**2./cos(p['Theta_in']*PI/180.)+1./cos(p['Theta_out']*PI/180.))

        sinalbessel=e*0
        sinalgaussiana=e*0

        # Calcula a contribuição de cada camada
        for x in arange(passox,espessura,passox): 

            # Seção de choque - CSRf(Z1, Z2, Ein, M1, M2, Theta):
            CS = c[int(x/passox)]*k*CSRf(ion['Z'], Edict[componente]['Z'], k*p['E0']-dedxef*x, mi, mt, Theta_s)*passox 
            # É multiplicada pelo passo para manter a escala do gráfico independente do mesmo.

            # Bessel:  
            # Ainda falta incluir a delta de Dirac na origem e a convolução com a resolução experimental
            # Existe um valor máximo em energia no qual a função de Bessel pode ser computada em função
            # da precisão numérica. Uma possível solução pode ser trabalhar em outras unidades que não
            # resultem em argumentos tão grandes para a função exponencial.         
            # Também temos um problema com a primeira camada, que tem profundidade zero.               
            # Perceba que os cálculos de Bessel correspondentes às camadas mais fundas estão truncados
            #  para uma perda de energia maior que 25 keV. (Observado com Theta_in=70, para x > 30)
            if modelo == 'bessel':
                alpha=dedxef*(2./dw2dxef)
                m=alpha*dedxef
                lbd=m*x*alpha
                besselcamada = lbd*exp(-m*x-alpha*(k*p['E0']-e))*i1(2.*sqrt(lbd*(k*p['E0']-e)))/(sqrt(lbd*(k*p['E0']-e)))
                sinalbessel = besselcamada*CS+sinalbessel

            # Gaussiana:
            elif modelo == 'gaussiana':
                Em=k*p['E0']-x*dedxef
                sigma=x*dw2dxef
                gaussianacamada = exp((-(e-Em)**2)/(2.*sigma))/sqrt(2.*PI*sigma) 
                sinalgaussiana = gaussianacamada*CS+sinalgaussiana

        if modelo == 'bessel':
            besself = nan_to_num(sinalbessel) 
            if LScontrol == 1:
                besself = convoluiF0(besself, alpha_lineshape, EPasso)
            Y = besself + Y
        elif modelo == 'gaussiana':
            if LScontrol == 1:     
                sinalgaussiana = convoluiF0(sinalgaussiana, alpha_lineshape, EPasso)           
            Y = sinalgaussiana + Y

    plt.plot(e,Y)
    suptitle('Signal received', fontsize=12)
    xlabel("Enegy (eV)")
    ylabel("Yield")

    amostra = np.loadtxt('H.py')
    xxx= (amostra[:,0]*15000/650)+85000
    ccc=amostra[:,1]*15e-37 -1e-34

    plot (xxx, ccc)

    show()

####################################################################################################################################################