xnx / pyvalem Goto Github PK

e.g. n=2, l=1 not l=1, n=2 for the "2p" configuration.

NB does J = # get interpreted as a rotational quantum number when fine-structure (total electronic angular momentum for an atom or ion) is intended?

Problem:

Currently, Formula('He+42') happily executes, while the maximal charge should be 2:

In [1]: Formula('He+42').charge
Out[1]: 42

Suggested solution:

Ensure the total atomic number of Formula instance is greater or equal than the charge:

In [47]: formula = Formula('CH4+42')
    ...: total_atomic_number =  sum(formula.atom_stoich[atom.symbol] * atom.Z for atom in formula.atoms)
    ...: if total_atomic_number < formula.charge:
    ...:     raise FormulaParseError(f'Charge {formula.charge} if greater than the total atomic number {total_atomic_number}!')
    ...: 
    ...: 
---------------------------------------------------------------------------
FormulaParseError                         Traceback (most recent call last)
<ipython-input-47-b8fa6fbad41d> in <module>
      2 total_atomic_number =  sum(formula.atom_stoich[str(atom)] * atom.Z for atom in formula.atoms)
      3 if total_atomic_number < formula.charge:
----> 4     raise FormulaParseError(f'Charge {formula.charge} if greater than the total atomic number {total_atomic_number}!')
      5 

FormulaParseError: Charge 42 if greater than the total atomic number 10!

Problems with the suggested solution:

• Some formulas do not define charge
• The mapping between the Atom and Isotope instances from Formula.atoms and their atom_symbols from Formula.atom_stoich.keys() is not defined by atom.symbol, but rather appears to be built up through a series of if ... else cases. However they appear to match.

Is there a way to express solid solution using chemical formula? By solid solution, I am referring to formula describing a crystal composed of different kinds of atoms occupying equivalent sites within the crystal lattice. This is a common occurrence for natural materials, i.e. minerals. I've included a few examples below:

olivine
Commonly expressed as (Mg, Fe)₂SiO₄, equivalent to (Mg_1−xFe_x)₂SiO₄. The first formulation is a very common method used by geochemists to denote solid solution where in this case Fe and Mg exist and can vary in nature from pure Mg to pure Fe, with any ratio of the two possible.

hornblende
This family of minerals is particularly complex, with numerous varieties existing, a product of numerous solid solution series existing:

• Magnesiohornblende-ferrohornblende, Ca₂[(Mg,Fe)₄Al]Si₇AlO₂₂(OH)₂
  • (Mg,Fe)₄ equivalent to (Mg_1-x,Fe_x)₄
• Tschermakite-ferrotschermakite, Ca₂[(Mg,Fe)₃Al₂]Si₆Al₂O₂₂(OH)₂
  • (Mg,Fe)₃ equivalent to (Mg_1-x,Fe_x)₃
• Edenite-ferroedenite, NaCa₂(Mg,Fe)₅Si₇AlO₂₂(OH)₂
  • (Mg,Fe)₅ equivalent to (Mg_1-x,Fe_x)₅
• Pargasite-ferropargasite, NaCa₂[(Mg,Fe)₄Al]Si₆Al₂O₂₂(OH)₂
  • (Mg,Fe)₄ equivalent to (Mg_1-x,Fe_x)₄
• Magnesiohastingstite-hastingsite, NaCa₂[(Mg,Fe)₄Fe3+]Si₆Al₂O₂₂(OH)₂
  • (Mg,Fe)₄ equivalent to (Mg_1-x,Fe_x)₄

Any thoughts as to feasibility of expressing such non-stoichiometric formulae? Thanks in advance!

I should note I borrowed heavily from wikipedia for the examples. Thanks Wikipedia editors!

Formulas like [Cu(H2O)4]SO4*H2O are usual in chemistry(see https://en.wikipedia.org/wiki/Glossary_of_chemical_formulae), but pyvalem seems to have problem with that. No formulas with nested parenthesis work for me. Is that a bug?

f = Formula('H20')#OK
f = Formula('[Cu(H2O)4]SO4*H2O') #fails
<...>
pyvalem.formula.FormulaParseError: Invalid formula syntax: [Cu(H2O)4]SO4*H2O
f = Formula('[Cu(H2O)4]SO4') #fails
<...>
pyvalem.formula.FormulaParseError: Invalid formula syntax: [Cu(H2O)4]SO4
f = Formula('(Cu(H2O)4)SO4') #fails
<...>
pyvalem.formula.FormulaParseError: Invalid formula syntax: (Cu(H2O)4)SO4

Unless I'm missing something in the underlying physics, I think states like '1v1 + 0v2 + 0v3' should be either parsed as '1v1', or raise an exception. This is what happens instead:

In [20]: StatefulSpecies('M 0v1+0v2+0v3')
Out[20]: M ν1+ν2+ν3

In [21]: StatefulSpecies('M 1v1+0v2+0v3')
Out[21]: M ν1+ν2+ν3

In [22]: StatefulSpecies('M 0v1+1v2+0v3')
Out[22]: M ν1+ν2+ν3

In [23]: StatefulSpecies('M 0v1+0v2+1v3')
Out[23]: M ν1+ν2+ν3

In [24]: StatefulSpecies('M 1v1+1v2+1v3')
Out[24]: M ν1+ν2+ν3