{"id":181,"date":"2016-12-09T16:10:40","date_gmt":"2016-12-09T21:10:40","guid":{"rendered":"https:\/\/people.clas.ufl.edu\/deumens\/?page_id=181"},"modified":"2026-03-19T08:36:39","modified_gmt":"2026-03-19T12:36:39","slug":"endyne","status":"publish","type":"page","link":"https:\/\/people.clas.ufl.edu\/deumens\/research\/endyne\/","title":{"rendered":"ENDyne"},"content":{"rendered":"\r\n<section class=\"fullwidth-text-block\">\r\n\t<div class=\"container px-0 pt-5\">\r\n\t\t<div class=\"row align-items-start\">\r\n\t\t\t<div class=\"col-12\">\r\n\t\t\t\t\n<h2 class=\"wp-block-heading\">Software for Dynamics of Electrons and Nuclei in Molecules.<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">Author: Erik Deumens.<\/h3>\n\n\n\n<p><span style=\"color: #ff0000\">ENDyne<\/span> is an application that implements the Electron Nuclear Dynamics (<span style=\"color: #ff0000\">END<\/span>) theory for studying the interaction between molecular geometry and electronic structure in a time-dependent and self-consistent way.<\/p>\n\n\n\n\n\n<p>The theory is somewhat unfamiliar to most people and the software is not very user friendly, for that reason we do not make the code available to a general audience at this time. However, if someone is interested in the code, they can come and study with us for about a month and they get to take the code with them at the end. Please send e-mail. for more information.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Program structure<\/h3>\n\n\n\n<pre>         ENDyne              Python\r\n         _______      _______________________\r\n         endmain      \"import endyne\"  PyRun_\r\n            |            |               ^\r\n            v            v               |             \r\n        ...............endyne.so.........|.......\r\n        .   \\ \\          |               |\t.\r\n        .    \\ v         v               |\t.\r\n        .     \\ pythonendcmds      pythonembed\t.\r\n        .      \\      |                ^\t.\r\n        ........\\.....|...............\/..........\r\n                 \\    |              \/\r\n                  \\   |             \/\r\n                   \\  |            \/\r\n                    v v           \/\r\n                   endynekernel.so\r\n<\/pre>\n\n\n\n\n\n\n\n<h3 class=\"wp-block-heading\">Feaures under development for version 5<\/h3>\n\n\n\n<ul class=\"wp-block-list\"><li>Multi configuration wave functions for electrons and nuclei<\/li>\n<li>Frozen electron cores<\/li>\n<li>Semi-empirical integrals<\/li>\n<li>Density Functional implementation<\/li>\n<li>Python programming interface<\/li>\n<li>Source in Fortran 95, ANSI C and python.<\/li>\n<li>MPI massively parallel<\/li>\n<li>POSIX threads shared memory parallel<\/li>\n<li>Use of <a href=\"http:\/\/users.clas.ufl.edu\/deumens\/qtip.html\">QTIP<\/a> integrals<\/li><\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><span style=\"color: #ff0000\">New features<\/span> in version 5<\/h3>\n\n\n\n\n\n\n\n<h3 class=\"wp-block-heading\">List of developers<\/h3>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr bgcolor=\"#33FF33\">\n<td>Author<\/td>\n<td>Project<\/td>\n<\/tr>\n<tr>\n<td>Trygve Helgaker<\/td>\n<td>McMurchie-Davisson gaussian orbital integrals<\/td>\n<\/tr>\n<tr>\n<td>Augie Diz<\/td>\n<td>SCF and state projection for diatomic<\/td>\n<\/tr>\n<tr>\n<td>Hugh Taylor<\/td>\n<td>vibration analysis for diatomic<\/td>\n<\/tr>\n<tr>\n<td>Benny Mogensen<\/td>\n<td>complex gaussian orbital integrals<\/td>\n<\/tr>\n<tr>\n<td>Jorge Morales<\/td>\n<td>differential cross sections with semiclassical corrections (Airy and uniform approximation)<\/td>\n<\/tr>\n<tr>\n<td>Mauricio Coutinho<\/td>\n<td>general state projection<\/td>\n<\/tr>\n<tr>\n<td>Anatol Blass<\/td>\n<td>general rotation-vibration analysis<\/td>\n<\/tr>\n<tr>\n<td>Remigio Trujillo<\/td>\n<td>differential cross sections with semiclassical corrections (Schiff approcimation); make movies with sphere representation of the Mulliken population.<\/td>\n<\/tr>\n<tr>\n<td>Denis Jacquemin<\/td>\n<td>Semi-empirical integrals (AM1) and Obara-Saika (PRISM) integrals.<\/td>\n<\/tr>\n<tr>\n<td>Oliver Quinet<\/td>\n<td>Python scripts for managing differential cross-section computations; make movies with true electron densities.<\/td>\n<\/tr>\n<tr>\n<td>Ben Hall<\/td>\n<td>Semi-classical nuclear wave function.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Python interface<\/span><\/td>\n<td width=\"85%\">The program <i>evolve<\/i> no longer exists in version 5. It is replaced by Python. <a href=\"http:\/\/www.python.org\">Python<\/a> is an Opensource project featuring a rapid prototying laguage that is both a simple scripting language and a fully featured object oriented language.ENDyne is defined as an extension of Python, which means that you can start the python interpreter and load ENDyne to make the ENDyne commands and data available in the interactive session or Python script. A session looks like:<!--PARAGRAPH_SEPARATOR--><pre>edsparc% python\r\nPython 2.0 (#3, Sep  3 2001, 11:50:37) [C] on sunos5\r\nType \"copyright\", \"credits\" or \"license\" for more information.\r\n&gt;&gt;&gt; import endyne\r\n&gt;&gt;&gt; endyne.version()\r\n(5, 'A', 3, 1, 6, 'generic', 'Tue Jul 30 12:52:28 EDT 2002')\r\n&gt;&gt;&gt; \r\n<\/pre>\n<p>Python is also embedded in the endyne executable, so that you can run endyne with a .inp file as argument to make a calculation as in version 2:<\/p><!--PARAGRAPH_SEPARATOR--><pre>edsparc% endyne test.inp\r\nENDyne version 5.A.3.1 test.inp\r\n\r\n                             *****************\r\n                             *  E N D y n e  *\r\n                             *****************\r\n\r\n                   Executable version  5 alpha development phase  3\r\n                           Restart file format  6\r\n\r\n<\/pre>\n<p>Or you can run endyne with a .rsl file as argument to get the functionality of <i>evolve<\/i>:<\/p><!--PARAGRAPH_SEPARATOR--><pre>edsparc% endyne test.rsl\r\nENDyne version 5.A.3.1 test.rsl\r\nRSL file test.rsl opened with the command:\r\n    rslfile = RSLopen('test.rsl','r')\r\nRecorded state log (RSL) file\r\n name:    test.rsl\r\n mode:    r\r\n format:   6\r\n version:      5.0603\r\n begin:    0.0E+0 1\r\n end:      0.1 29\r\n current:  0.0E+0 -1\r\n records:  11\r\n flags:   ownself       T\r\n          seekdone      T\r\n          readonly      T\r\n          flushbuf      F\r\n          isopen        T\r\n          rstisopen     F\r\n\r\nendyne&gt;&gt;&gt; \r\n<\/pre>\n<p>All the functionality of ENDyne can be accessed from Python and many of the new features are only available through the Python interface. Documentation of any command is available with the standard Python method:<\/p><!--PARAGRAPH_SEPARATOR--><pre>endyne&gt;&gt;&gt; print thing.__doc__\r\n<\/pre>\n<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Schiff Approximation<\/span><\/td>\n<td width=\"85%\">When a bundle of trajectories as a function of orientation angles of projectile and target and of impact parameters has been calculated, it is possible to get the differential cross section with the Python script <i>dcs.py<\/i>, which use the Python library <i>dcslib.py<\/i>. The semi-classical Schiff Approximation correction can be obtained with the Python function <i>schiff<\/i>.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Electronic subspace projection<\/span><\/td>\n<td width=\"85%\">The Mulliken population analysis provides away to estimate the probability of charge transfer. A more detailed way involves constructing the projection of the evolved state on the space of all determinants with given charge and spin distribution. This is done with the <i>project<\/i> method of <i>RSLdata<\/i> objects.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Thouless ordering<\/span><\/td>\n<td width=\"85%\">By setting <strong>LORDV2=T<\/strong>, the algorithm that determines the best reference for Thouless representation now takes into account the charge of each nucleus and builds a reference state that has a number of electrons on each nucleus in agreement with chemical intuition.To allow application to negative ions, the algorithm allows putting too many elecrons on a center, after all centers are chemically neutral.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Backstep recovery<\/span><\/td>\n<td width=\"85%\">Version 2 propagation relies on the RSL (recorded state log) file to recover from a problem where further integration is not possible. This happens because the integrator has propagated into a region of phase space where various operations to determine the forces on nuclei and electrons become numerically ill-conditioned. Continuing would lead to inaccurate and incorrect results. If the RSL step <strong>RDT<\/strong> is too large, the backstep may not yield a numerically more stable reference and then integration is aborted. This is known as a <i>broken trajectory<\/i>.Version 5 uses an internal mechanism to store the last known good state in RAM. This method is more flexible and requires little or no user intervention. It works effectively with the Hindmarsh-Gear <strong>LUSEHG=T<\/strong> and Shampine-Gordon <strong>LUSESG=T<\/strong> methods. Because the Bulirsch-Stoer <strong>LUSEBS=T<\/strong> method does not step forward uniformly, the RAM backstep method is really equivalent to the old RSL method.<!--PARAGRAPH_SEPARATOR--><p>The use of the parameter <strong>RCNDLM<\/strong> is much more sensitive in the new algorithm. Therefore users who have set this parameter, should remove it from their input files to ensure the use of the default. A new parameter <strong>RCNDAC<\/strong> has been introduced, giving the relative accuracy of the condition number returned by LAPACK routines. The version 2 algorithm considered a new state bad if<\/p><!--PARAGRAPH_SEPARATOR--><p><i>new_condition_number &gt; RCNDLM * NDIM * old_condition_number<\/i><br>\nindependent of how close in time the previous condition number was obtained. The version 5 algorithm uses the inequality<br>\n<i>d\/dt ln condition_number &gt; RCNDLM * NDIM<\/i><br>\nwithin the accuracy <i>RCNDAC<\/i>. Setting <strong>RCNDAC=1.<\/strong> and <strong>RCNDLM<\/strong> large ensures that the test is never satisfied. The resulting trajectory may then inaccurate and eventually wrong.<\/p><\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Project and target orientation<\/span><\/td>\n<td width=\"85%\">To compute differential cross sections for reactions involving non-spherical systems, it is necessary to compute sets of collision trajectories as a function of impact parameter for a set of orientations of the projectile and the target. The Python script <i>prep.py<\/i> will set up a directory tree and populate it with .inp files for <i>ENDyne<\/i> and .job files for <i>LoadLeveler<\/i> by editing tenplates provided by the user. The routines to average cross sections accurately over all orientations are provided in <i>orientlib.py<\/i>. These routines are used by <i>dcs.py<\/i> and <i>dcslib.py<\/i>.When both target and projectile are atoms, there are three relative coordinates:<br>\n1) b: the impact parameter along the b axis,<br>\n2) z: the initial projectile distance along the beam axis,<br>\n3) phi: the angle defining the b-axis beam-axis plane around the beam axis.<br>\nFor the most general case of two molecules, there are nine relative coordinates: b, z, phi as in the atom-atom case, plus:<br>\n4) alpha: azimuth for the projectile axis,<br>\n5) beta: azimuth for the target axis,<br>\n6) gamma: dihedral angle between the projectile axis and target axis,<br>\n7) chi: dihedral angle between the b-axis and beam-axis plane, and the projectile axis,<br>\n8) delta: projectile spin angle around the projectile axis,<br>\n9) epsilon: target spin angle around the target axis.There are 6 distinct types with increasing number of relative coordinates.<br>\n<strong>39<\/strong> The case of an atom and an atom has just three: b, z, phi.<br>\n<strong>59<\/strong> The case of an atom and a diatomic molecule has five: b, z, phi, beta, chi.<br>\n<strong>69<\/strong> The case of an atom and a polyatomic has six: b, z, phi, beta, chi, epsilon.<br>\n<strong>79<\/strong> The case of two diatomic molecules has seven: b, z, phi, alpha, beta, gamma, chi.<br>\n<strong>89<\/strong> The case of a diatomic molecule and a polyatomic has eight: b, z, phi, alpha, beta, gamma, chi, epsilon.<br>\n<strong>99<\/strong> The case of two polyatomics has all nine: b, z, phi, alpha, beta, gamma, chi, delta, epsilon.The program does not care whether the first cluster is the projectile or the target, hence to scatter a polyatomic on an atom, designate the atom as <i>fragment 1<\/i> and <i>target<\/i> and the polyatomic as <i>fragment 2<\/i> and <i>projectile<\/i> and use type <strong>69<\/strong>.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Performance Monitor<\/span><\/td>\n<td width=\"85%\">The program has been instrumented with calls to performance monitoring code. On all platform on which PAPI is supported, a package <i>hpm.c<\/i> provides a simple interface. This interface is the same as the one developed for IBM POWER3 and POWER4 processors by Luis DeRose called HPM Toolkit. On these platforms, ENDyne uses the HPM Toolkit. Monitoring is turned on with the flags <strong>LTRACE<\/strong> and <strong>LTRPRF<\/strong>.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Free electrons<\/span><\/td>\n<td width=\"85%\">It is possible to construct a single determinant wave function with complex orbitals using basis functions with electron translation factors (ETF) placed on centers that can be atomic nuclei or freely moving dynamical centers. We refer to such wave functions as wave functions <span style=\"color: #ff0000\">with free electrons<\/span>. The dynamical equations with dictate the motion of the free centers and of the occupation of the basis functions on them.Wave function with free electrons can be used to describe such processes as <span style=\"color: #ff0000\">electron scattering<\/span> on atoms and molecules;<span style=\"color: #ff0000\">ionization<\/span> of atoms and molecules during colissions and reactive processes; and <span style=\"color: #ff0000\">recombination<\/span> of ions with free electrons.<!--PARAGRAPH_SEPARATOR--><p>The use of the much more stringent backstep algorithm implemented in version 5 is essential for wave functions with free electrons, as the numerical integrator can produce nonsense much more easily than for wave functions with only atomic centers.<\/p><!--PARAGRAPH_SEPARATOR--><p>The use of coordinates for the center of free moving centers of basis functions introduces a duplication with the basis functions exapansion coefficients. To obtain a singularity free coordinate system for the phase space it is essential to perform a transformation. A new way to solve the dynamical equations has been implemented which constructs the complete symplectic form. The TDVP equations were solved by substitution in version 2, to save the space for the full symplectic form. However, the transformation is only meaningful on the entire symplectic form. The solution method with the full symplectic form is used when <strong>LUSPFM=T<\/strong>, and the singularity removing transformation is applied when <strong>LUSPTR=T<\/strong>. The Hamiltonian equation is solved with a general linear eqaution solver by default; with <strong>LUSPVD=T<\/strong> the equation is solved by singular value decomposition.<\/p><!--PARAGRAPH_SEPARATOR--><p>By turning on tracing <strong>LTRACE=T<\/strong> and <strong>LTRSPD=T<\/strong> detailed information about the solution of the equations is printed to the log file and, with <strong>LUSPVD=T<\/strong>, the singular values and right singular vectors are written to the RSL file in <strong>TRVEC1<\/strong> and <strong>TRMAT1<\/strong>, respectively. The components of the singular vectors are in the order of the internal representation of the symplectic equation: real part of the Thouless coefficients, center positions, imaginary part of the Thouless coefficients, center momenta. <strong>TRVEC2<\/strong> and <strong>TRVEC3<\/strong> contain the velocities as computed by the substitution method (version 2) and by solving the symplectic form (version 5, <strong>LUSPFM=T<\/strong>). <strong>TRVEC4<\/strong> holds the solution with the symplectic transformation <strong>LUSPTR=T<\/strong>. The velocity components are in the following order: Thouless coefficients, center positions, center momenta. <strong>TRVEC5<\/strong> contains the force of the TDVP equation. The force components are in the same order as the velocities.<\/p><!--PARAGRAPH_SEPARATOR--><p>There is an extra singularity that can develop: the free center can get zero coefficients for all its basis functions. The algorithm to avoid this singularity in the nuclear metric uses an expanded state phase space. This feature can be turned on by setting <strong>LUSEXS=T<\/strong>.<\/p><!--PARAGRAPH_SEPARATOR--><p>Rules for choosing and constructing the basis set for free centers are still being developed. So far the only guideline is that there should be only basis functions of even angular momentum: <i>s,d,g,j,&#8230;<\/i> and not <i>p,f,h,&#8230;<\/i>.<\/p><\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Optimization<\/span><\/td>\n<td width=\"85%\">The strength of ENDyne has always been in the time propagation. The capability to optimize electronic parameters at a fixed geometry, or to optimize electronic parameters and geometry simultaneously has always existed but did not perform very well. For that reason a conventional SCF code based on DIIS has been implemented to allow constructing initial SCF wave functions.The optimization capability has now been completely disentangled from the propagation capability, thus increasing the stability and maintainability of the software system. The optimization functionality appears to be much improved.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Phase normalization<\/span><\/td>\n<td width=\"85%\">Routines that diagonalize matrices, do not produce thesame vectors, they are determined only up to a complex phase. Now all Fock eigenvectors are normalized so that the largest component is real and positive.Reordering can produce different looking Thouless coefficientcs. Now the Thouless coefficients printed in the output at the end of a run are normalized so that the largest component is equal to 1. The speed vectors are similarly normalized, which makes no sense!<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">TDA and RPA spectra and modes<\/span><\/td>\n<td width=\"85%\">New flags <strong>LCIEIG<\/strong> and <strong>LCIVEC<\/strong> control that the configuration interaction singles or Tamm-Dancoff approximation energies and vectors will be computed with the current state as reference, which need not be an SCF state. Similarly, <strong>LIPEIG<\/strong> and <strong>LIPVEC<\/strong> control that the interaction picture or Random Phase approximation energies and vectors will be computed. These properties require the <strong>LMKPRP<\/strong> is set.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Interaction picture coordinates for propagation<\/span><\/td>\n<td width=\"85%\">The flag <strong>LUSEIP<\/strong> turns on the use of a transformation of the electronic variables to interaction picture variables, i.e. RPA modes, and propagation in these variables. This has the potential to significantly speed up propagation of electronic variables. The variables are monitored and when the forces get too big, indicating they are no longer optimal, evolution is stopped and a new transformation is forced before evolution is resumed.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">New computational kernel<\/span><\/td>\n<td width=\"85%\">The computational kernel has been rewritten completely. The old <strong>HFFRC<\/strong> routine with its complicated spin block structure has been broken up into smaller pieces with a simpler spin orbital structure more close in design to the structure required to support multi-configuration wave functions. The logic to determine a new mapping of basis functions to occupied\/unoccupied spaces needed to transform to Thouless coefficients has been modified to be closer to the fully general design needed for MC. A new input switch <strong>LMKTHC<\/strong> has been introduced to allow turning off generation of a new Thouless map, thus keeping the existing map, e.g. from the RSL file when <strong>LRSTRT=T<\/strong>.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Vector Hartree Fock<\/span><\/td>\n<td width=\"85%\">The Complete Active Space (CAS) multiconfiguration (MC) or multi reference (MR) (spin-)unrestricted vector coherent state wave function is being implemented and tested. It is more a generalization of Hartree-Fock than the conventional CAS-MC wave function. For that reason we call it the <strong>Vector Hartree-Fock<\/strong> wave function.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Semi-classical nuclear wave function<\/span><\/td>\n<td width=\"85%\">The nuclear de Broglie phase factor is integrated numerically alongthe trajectories and multiple trajectories are allowed to interfere. This is an implementation of the semi-classical nuclear wave function to generalize the Schiff approximation to the classical cross sections.<\/td>\n<\/tr>\n<tr>\n<td valign=\"top\" width=\"15%\"><span style=\"color: #ff0000\">Semi empirical integrals<\/span><\/td>\n<td width=\"85%\">The integral generation module has been completely restructured and now delivers semi empirical integrals and ab initio integrals in a consistent way. The code to compute contractions of integral patches with densities has been restructured as well. The new algorithm for evaluating the contractions is the same for MD model-forces, MNDO methods and ab initio methods.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<h3 class=\"wp-block-heading\">Software versions<\/h3>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr bgcolor=\"#33FF33\">\n<td width=\"8%\">Version<\/td>\n<td width=\"8%\">Release<\/td>\n<td width=\"8%\">Update<\/td>\n<td width=\"14%\">Date<\/td>\n<td width=\"60%\">Feature<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">2<\/td>\n<td width=\"8%\">7<\/td>\n<td width=\"8%\">6<\/td>\n<td width=\"14%\">Jan 1, 1998<\/td>\n<td width=\"60%\">Production; general molecules; electron translation factors<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr bgcolor=\"#33FF33\">\n<td width=\"8%\">Version<\/td>\n<td width=\"8%\">Release<\/td>\n<td width=\"8%\">Update<\/td>\n<td width=\"14%\">Date<\/td>\n<td width=\"60%\">Feature<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">2<\/td>\n<td width=\"8%\">8<\/td>\n<td width=\"8%\">1<\/td>\n<td width=\"14%\">Jan 1, 2000<\/td>\n<td width=\"60%\">Fortran 77; use QTIP integrals<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">2<\/td>\n<td width=\"8%\">8<\/td>\n<td width=\"8%\">1<\/td>\n<td width=\"14%\">Nov 1, 2000<\/td>\n<td width=\"60%\">Fortran 90 rewrite<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr bgcolor=\"#33FF33\">\n<td width=\"8%\">Version<\/td>\n<td width=\"8%\">Release<\/td>\n<td width=\"8%\">Phase<\/td>\n<td width=\"14%\">Date<\/td>\n<td width=\"60%\">Feature<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">5<\/td>\n<td width=\"8%\">Alpha<\/td>\n<td width=\"8%\">2<\/td>\n<td width=\"14%\">Mar 10, 2002<\/td>\n<td width=\"60%\">python integration; state projection; schiff approximation; use QTIP 1.A.4<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">5<\/td>\n<td width=\"8%\">Alpha<\/td>\n<td width=\"8%\">3<\/td>\n<td width=\"14%\">Jul 31, 2002<\/td>\n<td width=\"60%\">port to MIPS, Alpha, x86; use QTIP 1.A.4<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">5<\/td>\n<td width=\"8%\">Alpha<\/td>\n<td width=\"8%\">4<\/td>\n<td width=\"14%\">Dec 31, 2003<\/td>\n<td width=\"60%\">new spin loops; TDA and RPA; use QTIP 1.A.4<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">5<\/td>\n<td width=\"8%\">Alpha<\/td>\n<td width=\"8%\">6<\/td>\n<td width=\"14%\">Q3, 2008<\/td>\n<td width=\"60%\">Vector Hartree-Fock; semi-classical nuclear wave function; use QTIP 1.A.4<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">5<\/td>\n<td width=\"8%\">Alpha<\/td>\n<td width=\"8%\">5<\/td>\n<td width=\"14%\">Q1, 2009<\/td>\n<td width=\"60%\">semi empirical integrals; use QTIP 1.A.5<\/td>\n<\/tr>\n<tr bgcolor=\"#33FF33\">\n<td width=\"8%\">Version<\/td>\n<td width=\"8%\">Release<\/td>\n<td width=\"8%\">Update<\/td>\n<td width=\"14%\">Date<\/td>\n<td width=\"60%\">Feature<\/td>\n<\/tr>\n<tr>\n<td width=\"8%\">5<\/td>\n<td width=\"8%\">1<\/td>\n<td width=\"8%\">0<\/td>\n<td width=\"14%\">TBA<\/td>\n<td width=\"60%\">vector Hartree-Fock wave function; use QTIP 1.1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<h3 class=\"wp-block-heading\">Hardware and software platforms tested<\/h3>\n\n\n\n<p><a name=\"v2\"><\/a> <strong>ENDyne version 2<\/strong><\/p>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr bgcolor=\"#33FF33\">\n<td>Hardware<\/td>\n<td>OS<\/td>\n<td>Fortran 90<\/td>\n<\/tr>\n<tr>\n<td>SUN SPARC<\/td>\n<td>Solaris 7<\/td>\n<td>Forte 6.2<\/td>\n<\/tr>\n<tr>\n<td>IBM PowerPC<\/td>\n<td>AIX 4.3.3<\/td>\n<td>XLF 7.1<\/td>\n<\/tr>\n<tr>\n<td>SGI MIPS<\/td>\n<td>IRIX 6.5<\/td>\n<td>Compilers 7.3<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td>Linux RH ES 4.5<\/td>\n<td>Intel Fortran 9.1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<p><a name=\"v5\"><\/a> <strong>ENDyne version 5<\/strong><\/p>\n\n\n\n<table width=\"100%\">\n<tbody>\n<tr bgcolor=\"#33FF33\">\n<td>Hardware<\/td>\n<td>OS<\/td>\n<td>C<\/td>\n<td>Fortran 90<\/td>\n<td>Python<\/td>\n<\/tr>\n<tr>\n<td>SUN SPARC<\/td>\n<td>Solaris 7\/8\/9<\/td>\n<td>Forte 6.2<\/td>\n<td>Forte 6.2<\/td>\n<td>2.1<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td>AIX 5L 5.1<\/td>\n<td>VAC 6.0<\/td>\n<td>XLF 8.1<\/td>\n<td>2.1<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td>Linux RH ES 4.5<\/td>\n<td>gcc 3.3<\/td>\n<td>Intel Fortran 9.1<\/td>\n<td>2.1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n<h3 class=\"wp-block-heading\">Theoretical reference papers<\/h3>\n\n\n\n<ul class=\"wp-block-list\"><li><span style=\"color: #3333ff\">E. Deumens, Y. \u00d6hrn, and B. Weiner<\/span>,<br>\n<span style=\"color: #ff0000\"> Coherent state formulation of multiconfiguration states <\/span><br>\nJ. Math. Phys. 32, 1166-1175 (1991).<\/li>\n<li><span style=\"color: #3333ff\">E. Deumens, A. Diz, R. Longo and Y. \u00d6hrn<\/span>,<br>\n<span style=\"color: #ff0000\"> Time-dependent theoretical treatments of the dynamics of electrons and nuclei in molecular systems,<\/span><br>\nRev. of Mod. Phys., 66, 917-983 (1994).<\/li>\n<li><span style=\"color: #3333ff\">E. Deumens and Y. \u00d6hrn<\/span>,<br>\n<span style=\"color: #ff0000\">Wave function phase space: An approach to the dynamics of molecular systems<\/span><br>\nJ. Chem. Soc. Faraday Trans., <strong>93<\/strong>, 919 (1997).<\/li>\n<li><span style=\"color: #3333ff\">Y. \u00d6hrn and E. Deumens<\/span>,<br>\n<span style=\"color: #ff0000\">Towards an ab initio treatment of the time-dependent Schr\u00f6dinger equation of molecular systems<\/span>,<br>\nJ. Phys. Chem., A<strong>103<\/strong>, 9545 (1999).<\/li><\/ul>\n\n\n\r\n\t\t\t<\/div>\r\n\t\t<\/div>\r\n\t<\/div>\r\n<\/section>\r\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":685,"featured_media":0,"parent":22,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"featured_post":"","footnotes":"","_links_to":"","_links_to_target":""},"class_list":["post-181","page","type-page","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/pages\/181","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/users\/685"}],"replies":[{"embeddable":true,"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/comments?post=181"}],"version-history":[{"count":5,"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/pages\/181\/revisions"}],"predecessor-version":[{"id":435,"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/pages\/181\/revisions\/435"}],"up":[{"embeddable":true,"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/pages\/22"}],"wp:attachment":[{"href":"https:\/\/people.clas.ufl.edu\/deumens\/wp-json\/wp\/v2\/media?parent=181"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}