Source code for burnman.classes.composition

from __future__ import print_function
# This file is part of BurnMan - a thermoelastic and thermodynamic toolkit for
# the Earth and Planetary Sciences
# Copyright (C) 2012 - 2018 by the BurnMan team, released under the GNU
# GPL v2 or later.

import numpy as np
from scipy.optimize import nnls

from ..tools.chemistry import dictionarize_formula, formula_mass
from ..tools.misc import OrderedCounter


def composition_property(func):
    """
    Decorator @composition_property to be used for cached properties of compositions.

    To be used on function in Composition or derived classes that should be exposed
    as read-only properties that are cached.
    """
    class mat_obj():

        def __init__(self, func):
            self.func = func
            self.varname = self.func.__name__

        def get(self, obj):
            if not hasattr(obj, "_cached"):
                raise Exception("The composition_property decorator could not find class member _cached. "
                                "Did you forget to call Composition.__init__(self) in __init___?")
            cache_array = getattr(obj, "_cached")
            if self.varname not in cache_array:
                cache_array[self.varname] = self.func(obj)
            return cache_array[self.varname]

    return property(mat_obj(func).get, doc=func.__doc__)


[docs]def file_to_composition_list(fname, unit_type, normalize): """ Takes an input file with a specific format and returns a list of compositions (and associated comments) contained in that file. Parameters ---------- fname : string Path to ascii file containing composition data. Lines beginning with a hash are not read. The first read-line of the datafile contains a list of tab or space-separated components (e.g. FeO or SiO2), followed by the word Comment. Following lines are lists of floats with the amounts of each component. After the component amounts, the user can write anything they like in the Comment section. unit_type : 'weight' or 'molar' Specify whether the compositions in the file are given as weight or molar amounts. normalize : boolean If False, absolute numbers of moles/grams of component are stored, otherwise the component amounts of returned compositions will sum to one (until Composition.renormalize() is used). """ lines = list(filter(None, [line.rstrip('\n').split() for line in open(fname) if line[0] != '#'])) n_components = lines[0].index("Comment") components = lines[0][:n_components] comments = [line[n_components:] for line in lines[1:]] compositions = np.array([map(float, l) for l in list(zip(*(list(zip(*lines[1:]))[:n_components])))]) return [Composition(OrderedCounter(dict(zip(components, c))), unit_type, normalize) for c in compositions], comments
[docs]class Composition(object): """ Class for a composition object, which can be used to store, modify and renormalize compositions, and also convert between molar, weight, and atomic amounts. This class is available as ``burnman.Composition``. """ def __init__(self, composition_dictionary, unit_type='weight', normalize=False): """ Create a composition using a dictionary and unit type. Parameters ---------- composition_dictionary : dictionary Dictionary of components (given as a string) and their amounts. unit_type : 'weight' or 'molar' (optional, 'weight' as standard) Specify whether the input composition is given as weight or molar amounts. normalize : boolean If False, absolute numbers of moles/grams of component are stored, otherwise the component amounts of returned compositions will sum to one (until Composition.renormalize() is used). """ self._cached = {} n_total = float(sum(composition_dictionary.values())) normalized_dictionary = {} for k in composition_dictionary.keys(): normalized_dictionary[k] = composition_dictionary[k]/n_total self.normalization_component = {'weight': 'total', 'molar': 'total', 'atomic': 'total'} self.normalization_amount = {'weight': 1., 'molar': 1., 'atomic': 1.} # component formulae self.component_formulae = {c: dictionarize_formula(c) for c in composition_dictionary.keys()} # elemental compositions of components self.element_list = OrderedCounter() for component in self.component_formulae.values(): self.element_list += OrderedCounter({element: n_atoms for (element, n_atoms) in component.items()}) self.element_list = list(self.element_list.keys()) if unit_type == 'weight': if normalize: self._cached['weight_composition'] = OrderedCounter(normalized_dictionary) else: self._cached['weight_composition'] = OrderedCounter(composition_dictionary) mole_total = sum([composition_dictionary[c] / formula_mass(self.component_formulae[c]) for c in composition_dictionary.keys()]) self.normalization_amount['weight'] = n_total self.normalization_amount['molar'] = mole_total self.normalization_amount['atomic'] = sum(self._moles_component_to_atoms(self.molar_composition).values()) elif unit_type == 'molar': if normalize: self._cached['molar_composition'] = OrderedCounter(normalized_dictionary) else: self._cached['molar_composition'] = OrderedCounter(composition_dictionary) weight_total = sum([composition_dictionary[c] * formula_mass(self.component_formulae[c]) for c in composition_dictionary.keys()]) self.normalization_amount['weight'] = weight_total self.normalization_amount['molar'] = n_total self.normalization_amount['atomic'] = sum(self._moles_component_to_atoms(self.molar_composition).values()) else: raise Exception('Unit type not yet implemented. ' 'Should be either weight or molar.')
[docs] def renormalize(self, unit_type, normalization_component, normalization_amount): """ Change the normalization for a given unit type (weight, molar, or atomic) Resets cached composition only for that unit type Parameters ---------- unit_type : 'weight', 'molar' or 'atomic' Unit type composition to be renormalised normalization_component: string Component/element on which to renormalize. String must either be one of the components/elements already in composite, or have the value 'total' normalization_amount: float Amount of component in the renormalised composition """ if unit_type == 'weight': s = self.weight_composition elif unit_type == 'molar': s = self.molar_composition elif unit_type == 'atomic': s = self.atomic_composition else: raise Exception('Unit type not recognised. ' 'Should be one of weight, molar and atomic') self.normalization_component[unit_type] = normalization_component self.normalization_amount[unit_type] = normalization_amount self._cached[unit_type+'_composition'] = self._normalize_to_basis(s, unit_type)
[docs] def add_components(self, composition_dictionary, unit_type): """ Add (or remove) components from the composition. The components are added to the current state of the (weight or molar) composition; if the composition has been renormalised, then this should be taken into account. Parameters ---------- composition_dictionary : dictionary Components to add, and their amounts, in dictionary form unit_type : 'weight' or 'molar' Unit type of the components to be added """ if unit_type == 'weight': composition = self.weight_composition elif unit_type == 'molar': composition = self.molar_composition else: raise Exception('Unit type not recognised. ' 'Should be either weight or molar.') composition += OrderedCounter(composition_dictionary) self.__init__(composition, unit_type)
[docs] def change_component_set(self, new_component_list): """ Change the set of basis components without changing the bulk composition. Will raise an exception if the new component set is invalid for the given composition. Parameters ---------- new_component_list : list of strings New set of basis components. """ composition = np.array([self.atomic_composition[element] for element in self.element_list]) component_matrix = np.zeros((len(new_component_list), len(self.element_list))) for i, component in enumerate(new_component_list): formula = dictionarize_formula(component) for element, n_atoms in formula.items(): component_matrix[i][self.element_list.index(element)] = n_atoms sol = nnls(component_matrix.T, composition) if sol[1] < 1.e-12: component_amounts = sol[0] else: raise Exception('Failed to change component set. ' 'Could not find a non-negative least squares solution. ' 'Can the bulk composition be described with this set of components?') composition = OrderedCounter(dict(zip(new_component_list, component_amounts))) self.__init__(composition, 'molar')
def _normalize_to_basis(self, composition, unit_type): if self.normalization_component[unit_type] == 'total': n_orig = float(sum(composition.values())) else: n_orig = composition[self.normalization_component[unit_type]] for k in composition.keys(): composition[k] *= self.normalization_amount[unit_type]/n_orig return composition @composition_property def molar_composition(self): """ Returns the molar composition as a counter [moles] """ mole_compositions = OrderedCounter({c: self.weight_composition[c] / formula_mass(self.component_formulae[c]) for c in self.weight_composition.keys()}) return self._normalize_to_basis(mole_compositions, 'molar') @composition_property def weight_composition(self): """ Returns the weight composition as a counter [g] """ weight_compositions = OrderedCounter({c: self.molar_composition[c] * formula_mass(self.component_formulae[c]) for c in self.molar_composition.keys()}) return self._normalize_to_basis(weight_compositions, 'weight') @composition_property def atomic_composition(self): """ Returns the atomic composition as a counter [moles] """ atom_compositions = self._moles_component_to_atoms(self.molar_composition) return self._normalize_to_basis(atom_compositions, 'atomic') def _moles_component_to_atoms(self, molar_composition_dictionary): component_matrix = np.zeros((len(self.component_formulae), len(self.element_list))) molar_composition_vector = np.zeros(len(self.component_formulae)) for i, (component, formula) in enumerate(self.component_formulae.items()): molar_composition_vector[i] = molar_composition_dictionary[component] for element, n_atoms in formula.items(): component_matrix[i][self.element_list.index(element)] = n_atoms atom_compositions = np.dot(molar_composition_vector, component_matrix) return OrderedCounter(dict(zip(self.element_list, atom_compositions)))
[docs] def print(self, unit_type, significant_figures=1, normalization_component='total', normalization_amount=100.): """ Pretty-print function for the composition This does not renormalize the Composition internally Parameters ---------- unit_type : 'weight', 'molar' or 'atomic' Unit type in which to print the composition significant_figures : integer Number of significant figures for each amount normalization_component: string Component/element on which to renormalize. String must either be one of the components/elements already in composite, or have the value 'total'. (default = 'total') normalization_amount: float Amount of component in the renormalised composition. (default = '100.') """ if unit_type == 'weight': print('Weight composition') if normalization_component == 'total': total_stored = float(sum(self.weight_composition.values())) else: total_stored = self.weight_composition[normalization_component] f = normalization_amount/total_stored for (key, value) in sorted(self.weight_composition.items()): print('{0}: {1:0.{sf}f}'.format(key, value*f, sf=significant_figures)) elif unit_type == 'molar': print('Molar composition') if normalization_component == 'total': total_stored = float(sum(self.molar_composition.values())) else: total_stored = self.molar_composition[normalization_component] f = normalization_amount/total_stored for (key, value) in sorted(self.molar_composition.items()): print('{0}: {1:0.{sf}f}'.format(key, value*f, sf=significant_figures)) elif unit_type == 'atomic': print('Atomic composition') if normalization_component == 'total': total_stored = float(sum(self.atomic_composition.values())) else: total_stored = self.atomic_composition[normalization_component] f = normalization_amount/total_stored for (key, value) in sorted(self.atomic_composition.items()): print('{0}: {1:0.{sf}f}'.format(key, value*f, sf=significant_figures)) else: raise Exception('unit_type not yet implemented. Should be either weight, molar or atomic.')