109 |
|
|
110 |
|
// use a small angle aproximation for sin and cosine |
111 |
|
|
112 |
< |
//angleSqr = angle * angle; |
113 |
< |
//angleSqrOver4 = angleSqr / 4.0; |
114 |
< |
//top = 1.0 - angleSqrOver4; |
115 |
< |
//bottom = 1.0 + angleSqrOver4; |
112 |
> |
angleSqr = angle * angle; |
113 |
> |
angleSqrOver4 = angleSqr / 4.0; |
114 |
> |
top = 1.0 - angleSqrOver4; |
115 |
> |
bottom = 1.0 + angleSqrOver4; |
116 |
|
|
117 |
< |
//cosAngle = top / bottom; |
118 |
< |
//sinAngle = angle / bottom; |
119 |
< |
cosAngle = cos(angle); |
120 |
< |
sinAngle = sin(angle); |
117 |
> |
cosAngle = top / bottom; |
118 |
> |
sinAngle = angle / bottom; |
119 |
> |
|
120 |
> |
// or don't use the small angle approximation: |
121 |
> |
//cosAngle = cos(angle); |
122 |
> |
//sinAngle = sin(angle); |
123 |
|
rot(axes1, axes1) = cosAngle; |
124 |
|
rot(axes2, axes2) = cosAngle; |
125 |
|
|
129 |
|
// rotate the momentum acoording to: ji[] = rot[][] * ji[] |
130 |
|
ji = rot * ji; |
131 |
|
|
132 |
< |
// rotate the Rotation matrix acording to: |
133 |
< |
// A[][] = A[][] * transpose(rot[][]) |
134 |
< |
// transpose(A[][]) = transpose(A[][]) * transpose(rot[][]) |
132 |
> |
// This code comes from converting an algorithm detailed in |
133 |
> |
// J. Chem. Phys. 107 (15), pp. 5840-5851 by Dullweber, |
134 |
> |
// Leimkuhler and McLachlan (DLM) for use in our code. |
135 |
> |
// In Appendix A, the DLM paper has the change to the rotation |
136 |
> |
// matrix as: Q = Q * rot.transpose(), but our rotation matrix |
137 |
> |
// A is actually equivalent to Q.transpose(). This fact can be |
138 |
> |
// seen on page 5849 of the DLM paper where a lab frame |
139 |
> |
// dipole \mu_i(t) is expressed in terms of a body-fixed |
140 |
> |
// reference orientation, \bar{\mu_i} and the rotation matrix, Q: |
141 |
> |
// \mu_i(t) = Q * \bar{\mu_i} |
142 |
> |
// Our code computes lab frame vectors from body-fixed reference |
143 |
> |
// vectors using: |
144 |
> |
// v_{lab} = A.transpose() * v_{body} |
145 |
> |
// (See StuntDouble.hpp for confirmation of this fact). |
146 |
> |
// |
147 |
> |
// So, using the identity: |
148 |
> |
// (A * B).transpose() = B.transpose() * A.transpose(), we |
149 |
> |
// get the equivalent of Q = Q * rot.transpose() for our code to be: |
150 |
|
|
151 |
< |
A = rot * A; //? A = A* rot.transpose(); |
151 |
> |
A = rot * A; |
152 |
|
|
153 |
|
} |
154 |
|
|