|
From: Dominik K. <dom...@gm...> - 2017-02-09 20:04:00
|
xrayutilities 1.4.0 was released (2017-02-09) During the last month several important improvements, changes, and additions were made to xrayutilities. As usual every user is advised to upgrade. In order to obtain the new version go to sourceforge [1] or use PyPI [2]. Windows binaries and wheel packages are found on PyPI for 64-bit Python in versions 2.7 and 3.5 and 3.6. With the change to version 1.4.0 there might be some code changes in your scripts required. Please check the list of changes below and see if you are likely to be affected. New simulation capabilities were added to xrayutilities. Notable are a dynamical multibeam formalism for symmetric crystal truncation rods and powder diffraction code based on the fundamental parameters approach. The core of the powder diffraction code was submitted by Marcus H. Mendenhall < mar...@ni...> and is documented in Journal of Research of NIST 120, 223 (2015) http://dx.doi.org/10.6028/jres.120.014 . Example scripts for powder diffraction (simpack_powdermodel.py) and the new dynamical diffraction simulation (simpack_xrd_Darwin_AlGaAs.py) can be found in the example folder [3]. Further changes include a complete redesign of how materials or better their lattices are defined. The new way of defining materials is more flexible and closer to the way most other software treats lattices. A crystalline material is now specified by its space group number and the Wyckoff position of atoms within the unit cell. For example a zincblende material one defines by: import xrayutilities as xu InP = xu.materials.Crystal("InP", xu.materials.SGLattice(216, 5.8687, atoms=[xu.materials.elements.In, xu.materials.elements.P], pos=['4a', '4c'])) InAs = xu.materials.Crystal("InAs", xu.materials.SGLattice(216, 6.0583, atoms=['In', 'As'], pos=['4a', '4c'])) The definition of the lattice takes a variable number of parameters depending on the symmetry of the given space group. A more complex definition is for example used for ferroelectric GeTe which is rhombohedral and its atomic positions have several degrees of freedom which need to be specified: GeTe = Crystal("GeTe", SGLattice('160:R', 5.996, 88.18, atoms=[e.Ge, e.Ge, e.Te, e.Te], pos=[('1a', -0.237), ('3b', (0.5-0.237, -0.237)), ('1a', 0.237), ('3b', (0.5+0.237, +0.237))])) Many more examples of material definitions can be found seen in the list of predefined materials [4]. With the new changes the definition of alloyed materials is much more straight forward since alloying is now automatically performed also within the unit cell. The only requirement is that both materials have the same space group! To create an InAs_{1-x} P_x alloy whose lattice parameter follows Vegard's rule one uses: InAsP = xu.materials.Alloy(InAs, InP, 0.4) print(InAsP) Alloy: InAs(0.60)InP(0.40) Lattice: 216 cubic F-43m: a = 5.9825, b = 5.9825 c= 5.9825 alpha = 90.000, beta = 90.000, gamma = 90.000 Lattice base: 0: In (49) 4a occ=0.600 b=0.000 1: As (33) 4c occ=0.600 b=0.000 2: In (49) 4a occ=0.400 b=0.000 3: P (15) 4c occ=0.400 b=0.000 Currently the old Lattice code is still included and can be used to make a primitive conversion of your old materials and help you convert your material definitions. If you for example had defined your own zincblende lattice/material you can convert it with: zb = xu.materials.ZincBlendeLattice(xu.materials.elements.In, xu.materials.elements.P, 5.8687) new_zb = xu.materials.SGLattice.fromLattice(zb) print(new_zb) ... As you will see from the output this conversion always uses the P1 space-group. This conversion code and all old Lattice definitions will be removed in the next major release. For a more exhaustive list of changes in version 1.4.0 see the CHANGES.txt file or the GIT log. Test coverage of this version is at 60%. [1] https://sourceforge.net/projects/xrayutilities [2] https://pypi.python.org/pypi/xrayutilities [3] https://github.com/dkriegner/xrayutilities/tree/master/examples [4] https://github.com/dkriegner/xrayutilities/blob/master/xrayutilities/materials/predefined_materials.py |
