1 | # ellipsoid model |
---|
2 | # Note: model title and parameter table are inserted automatically |
---|
3 | r""" |
---|
4 | The form factor is normalized by the particle volume |
---|
5 | |
---|
6 | Definition |
---|
7 | ---------- |
---|
8 | |
---|
9 | The output of the 2D scattering intensity function for oriented ellipsoids |
---|
10 | is given by (Feigin, 1987) |
---|
11 | |
---|
12 | .. math:: |
---|
13 | |
---|
14 | P(q,\alpha) = \frac{\text{scale}}{V} F^2(q,\alpha) + \text{background} |
---|
15 | |
---|
16 | where |
---|
17 | |
---|
18 | .. math:: |
---|
19 | |
---|
20 | F(q,\alpha) = \Delta \rho V \frac{3(\sin qr - qr \cos qr)}{(qr)^3} |
---|
21 | |
---|
22 | for |
---|
23 | |
---|
24 | .. math:: |
---|
25 | |
---|
26 | r = \left[ R_e^2 \sin^2 \alpha + R_p^2 \cos^2 \alpha \right]^{1/2} |
---|
27 | |
---|
28 | |
---|
29 | $\alpha$ is the angle between the axis of the ellipsoid and $\vec q$, |
---|
30 | $V = (4/3)\pi R_pR_e^2$ is the volume of the ellipsoid, $R_p$ is the polar |
---|
31 | radius along the rotational axis of the ellipsoid, $R_e$ is the equatorial |
---|
32 | radius perpendicular to the rotational axis of the ellipsoid and |
---|
33 | $\Delta \rho$ (contrast) is the scattering length density difference between |
---|
34 | the scatterer and the solvent. |
---|
35 | |
---|
36 | For randomly oriented particles use the orientational average, |
---|
37 | |
---|
38 | .. math:: |
---|
39 | |
---|
40 | \langle F^2(q) \rangle = \int_{0}^{\pi/2}{F^2(q,\alpha)\sin(\alpha)\,d\alpha} |
---|
41 | |
---|
42 | |
---|
43 | computed via substitution of $u=\sin(\alpha)$, $du=\cos(\alpha)\,d\alpha$ as |
---|
44 | |
---|
45 | .. math:: |
---|
46 | |
---|
47 | \langle F^2(q) \rangle = \int_0^1{F^2(q, u)\,du} |
---|
48 | |
---|
49 | with |
---|
50 | |
---|
51 | .. math:: |
---|
52 | |
---|
53 | r = R_e \left[ 1 + u^2\left(R_p^2/R_e^2 - 1\right)\right]^{1/2} |
---|
54 | |
---|
55 | For 2d data from oriented ellipsoids the direction of the rotation axis of |
---|
56 | the ellipsoid is defined using two angles $\theta$ and $\phi$ as for the |
---|
57 | :ref:`cylinder orientation figure <cylinder-angle-definition>`. |
---|
58 | For the ellipsoid, $\theta$ is the angle between the rotational axis |
---|
59 | and the $z$ -axis in the $xz$ plane followed by a rotation by $\phi$ |
---|
60 | in the $xy$ plane, for further details of the calculation and angular |
---|
61 | dispersions see :ref:`orientation` . |
---|
62 | |
---|
63 | NB: The 2nd virial coefficient of the solid ellipsoid is calculated based |
---|
64 | on the $R_p$ and $R_e$ values, and used as the effective radius for |
---|
65 | $S(q)$ when $P(q) \cdot S(q)$ is applied. |
---|
66 | |
---|
67 | |
---|
68 | The $\theta$ and $\phi$ parameters are not used for the 1D output. |
---|
69 | |
---|
70 | |
---|
71 | |
---|
72 | Validation |
---|
73 | ---------- |
---|
74 | |
---|
75 | Validation of the code was done by comparing the output of the 1D model |
---|
76 | to the output of the software provided by the NIST (Kline, 2006). |
---|
77 | |
---|
78 | The implementation of the intensity for fully oriented ellipsoids was |
---|
79 | validated by averaging the 2D output using a uniform distribution |
---|
80 | $p(\theta,\phi) = 1.0$ and comparing with the output of the 1D calculation. |
---|
81 | |
---|
82 | |
---|
83 | .. _ellipsoid-comparison-2d: |
---|
84 | |
---|
85 | .. figure:: img/ellipsoid_comparison_2d.jpg |
---|
86 | |
---|
87 | Comparison of the intensity for uniformly distributed ellipsoids |
---|
88 | calculated from our 2D model and the intensity from the NIST SANS |
---|
89 | analysis software. The parameters used were: *scale* = 1.0, |
---|
90 | *radius_polar* = 20 |Ang|, *radius_equatorial* = 400 |Ang|, |
---|
91 | *contrast* = 3e-6 |Ang^-2|, and *background* = 0.0 |cm^-1|. |
---|
92 | |
---|
93 | The discrepancy above $q$ = 0.3 |cm^-1| is due to the way the form factors |
---|
94 | are calculated in the c-library provided by NIST. A numerical integration |
---|
95 | has to be performed to obtain $P(q)$ for randomly oriented particles. |
---|
96 | The NIST software performs that integration with a 76-point Gaussian |
---|
97 | quadrature rule, which will become imprecise at high $q$ where the amplitude |
---|
98 | varies quickly as a function of $q$. The SasView result shown has been |
---|
99 | obtained by summing over 501 equidistant points. Our result was found |
---|
100 | to be stable over the range of $q$ shown for a number of points higher |
---|
101 | than 500. |
---|
102 | |
---|
103 | Model was also tested against the triaxial ellipsoid model with equal major |
---|
104 | and minor equatorial radii. It is also consistent with the cyclinder model |
---|
105 | with polar radius equal to length and equatorial radius equal to radius. |
---|
106 | |
---|
107 | References |
---|
108 | ---------- |
---|
109 | |
---|
110 | L A Feigin and D I Svergun. |
---|
111 | *Structure Analysis by Small-Angle X-Ray and Neutron Scattering*, |
---|
112 | Plenum Press, New York, 1987. |
---|
113 | |
---|
114 | A. Isihara. J. Chem. Phys. 18(1950) 1446-1449 |
---|
115 | |
---|
116 | Authorship and Verification |
---|
117 | ---------------------------- |
---|
118 | |
---|
119 | * **Author:** NIST IGOR/DANSE **Date:** pre 2010 |
---|
120 | * **Converted to sasmodels by:** Helen Park **Date:** July 9, 2014 |
---|
121 | * **Last Modified by:** Paul Kienzle **Date:** March 22, 2017 |
---|
122 | """ |
---|
123 | from __future__ import division |
---|
124 | |
---|
125 | import numpy as np |
---|
126 | from numpy import inf, sin, cos, pi |
---|
127 | |
---|
128 | try: |
---|
129 | from numpy import cbrt |
---|
130 | except ImportError: |
---|
131 | def cbrt(x): return x ** (1.0/3.0) |
---|
132 | |
---|
133 | name = "ellipsoid" |
---|
134 | title = "Ellipsoid of revolution with uniform scattering length density." |
---|
135 | |
---|
136 | description = """\ |
---|
137 | P(q.alpha)= scale*f(q)^2 + background, where f(q)= 3*(sld |
---|
138 | - sld_solvent)*V*[sin(q*r(Rp,Re,alpha)) |
---|
139 | -q*r*cos(qr(Rp,Re,alpha))] |
---|
140 | /[qr(Rp,Re,alpha)]^3" |
---|
141 | |
---|
142 | r(Rp,Re,alpha)= [Re^(2)*(sin(alpha))^2 |
---|
143 | + Rp^(2)*(cos(alpha))^2]^(1/2) |
---|
144 | |
---|
145 | sld: SLD of the ellipsoid |
---|
146 | sld_solvent: SLD of the solvent |
---|
147 | V: volume of the ellipsoid |
---|
148 | Rp: polar radius of the ellipsoid |
---|
149 | Re: equatorial radius of the ellipsoid |
---|
150 | """ |
---|
151 | category = "shape:ellipsoid" |
---|
152 | |
---|
153 | # ["name", "units", default, [lower, upper], "type","description"], |
---|
154 | parameters = [["sld", "1e-6/Ang^2", 4, [-inf, inf], "sld", |
---|
155 | "Ellipsoid scattering length density"], |
---|
156 | ["sld_solvent", "1e-6/Ang^2", 1, [-inf, inf], "sld", |
---|
157 | "Solvent scattering length density"], |
---|
158 | ["radius_polar", "Ang", 20, [0, inf], "volume", |
---|
159 | "Polar radius"], |
---|
160 | ["radius_equatorial", "Ang", 400, [0, inf], "volume", |
---|
161 | "Equatorial radius"], |
---|
162 | ["theta", "degrees", 60, [-360, 360], "orientation", |
---|
163 | "ellipsoid axis to beam angle"], |
---|
164 | ["phi", "degrees", 60, [-360, 360], "orientation", |
---|
165 | "rotation about beam"], |
---|
166 | ] |
---|
167 | |
---|
168 | |
---|
169 | source = ["lib/sas_3j1x_x.c", "lib/gauss76.c", "ellipsoid.c"] |
---|
170 | have_Fq = True |
---|
171 | effective_radius_type = ["equivalent sphere","average curvature", "min radius", "max radius"] |
---|
172 | |
---|
173 | def random(): |
---|
174 | volume = 10**np.random.uniform(5, 12) |
---|
175 | radius_polar = 10**np.random.uniform(1.3, 4) |
---|
176 | radius_equatorial = np.sqrt(volume/radius_polar) # ignore 4/3 pi |
---|
177 | pars = dict( |
---|
178 | #background=0, sld=0, sld_solvent=1, |
---|
179 | radius_polar=radius_polar, |
---|
180 | radius_equatorial=radius_equatorial, |
---|
181 | ) |
---|
182 | return pars |
---|
183 | |
---|
184 | demo = dict(scale=1, background=0, |
---|
185 | sld=6, sld_solvent=1, |
---|
186 | radius_polar=50, radius_equatorial=30, |
---|
187 | theta=30, phi=15, |
---|
188 | radius_polar_pd=.2, radius_polar_pd_n=15, |
---|
189 | radius_equatorial_pd=.2, radius_equatorial_pd_n=15, |
---|
190 | theta_pd=15, theta_pd_n=45, |
---|
191 | phi_pd=15, phi_pd_n=1) |
---|
192 | q = 0.1 |
---|
193 | # april 6 2017, rkh add unit tests, NOT compared with any other calc method, assume correct! |
---|
194 | qx = q*cos(pi/6.0) |
---|
195 | qy = q*sin(pi/6.0) |
---|
196 | tests = [ |
---|
197 | [{}, 0.05, 54.8525847025], |
---|
198 | [{'theta':80., 'phi':10.}, (qx, qy), 1.74134670026], |
---|
199 | ] |
---|
200 | del qx, qy # not necessary to delete, but cleaner |
---|