A ```fluid``` object can have multiple compositions. A multidimensional ```fluid``` object can be created for example by multiplication with a numpy array:
``` python
fl2 = gp.fluid({'H2O': 1, 'N2': 2}) + \
np.linspace(0, 10, 4) * gp.fluid({'H2': 1})
fl2
```
```
Total mol:
array([ 3. , 6.33333333, 9.66666667, 13. ])
Species:
H2 H2O N2
Molar fractions:
array([[0. , 0.33333333, 0.66666667],
[0.52631579, 0.15789474, 0.31578947],
[0.68965517, 0.10344828, 0.20689655],
[0.76923077, 0.07692308, 0.15384615]])
```
A fluid object can be converted to a pandas dataframe:
``` python
import pandas as pd
pd.DataFrame(list(fl2))
```
| | H2O | N2 | H2
|----|-----|-----|-------
|0 | 1.0 | 2.0 | 0.000000
|1 | 1.0 | 2.0 | 3.333333
|2 | 1.0 | 2.0 | 6.666667
|3 | 1.0 | 2.0 | 10.000000
The broadcasting behavior is not limited to 1D-arrays:
In some cases not the molecular but the atomic composition is of interest. The ```elements``` class can be used for atom based balances and works similar:
``` python
el = gp.elements({'N': 1, 'Cl': 2})
el.get_mass()
```
```
np.float64(0.08490700000000001)
```
A ```elements``` object can be as well instantiated from a ```fluid``` object. Arithmetic operations between ```elements``` and ```fluid``` result in an ```elements``` object:
``` python
el2 = gp.elements(fl) + el - 0.3 * fl
el2
```
```
Cl 2.000e+00 mol
H 4.200e+00 mol
N 1.000e+00 mol
O 7.000e-01 mol
```
Going from an atomic composition to an molecular composition is a little bit less straight forward, since there is no universal approach. One way is to calculate the thermodynamic equilibrium for a mixture:
``` python
fs = gp.fluid_system('CH4, H2, CO, CO2, O2')
el3 = gp.elements({'C': 1, 'H': 2, 'O':1}, fs)
fl3 = gp.equilibrium(el3, t=800)
fl3
```
```
Total 1.204e+00 mol
CH4 33.07 %
H2 16.93 %
CO 16.93 %
CO2 33.07 %
O2 0.00 %
```
The ```equilibrium``` function can be called with a ```fluid``` or ```elements``` object as first argument. ```fluid``` and ```elements``` referencing a ```fluid_system``` object witch can be be set as shown above during the object instantiation. If not provided, a new one will be created automatically. Providing a ```fluid_system``` gives more control over which molecular species are included in derived ```fluid``` objects. Furthermore arithmetic operations between objects with the same ```fluid_system``` are potentially faster:
``` python
fl3 + gp.fluid({'CH4': 1}, fs)
```
```
Total 2.204e+00 mol
CH4 63.44 %
H2 9.24 %
CO 9.24 %
CO2 18.07 %
O2 0.00 %
```
Especially if the ```fluid_system``` of one of the operants has not a subset of molecular species of the other ```fluid_system``` a new ```fluid_system``` will be created for the operation which might degrade performance:
