Negative molar fractions appear when using ReactorSurface #1880
Description
Problem description
When using ReactorSurface, negative concentrations appear.
Steps to reproduce
Run attached reproducer.py with attached reproducer.yaml. Essential parts are:
INITIAL_COMPOSITION = "H2:1,HN2+:0.01,N2:0.01"
file_name = "reproducer.yaml"
print(f"Loading mechanism '{file_name}' ...")
gas = ct.Solution(file_name, "gas")
print("Setting initial conditions ...")
gas.TPX = 300.0, 5, INITIAL_COMPOSITION
print("Creating reactor ...")
reactor = ct.IdealGasReactor(gas, volume=1.0, energy="off")
print("Creating surface ...")
surface_kin = ct.Interface(file_name, "surface")
surface = ct.ReactorSurface(surface_kin, reactor, A=0.1)
print(reactor.surfaces)
print("Creating network ...")
sim = ct.ReactorNet([reactor])
sim.verbose = True
reactor.thermo()
print("Integrating ...")
t_min = 1e-07
t_max = 1
time = np.geomspace(t_min, t_max, int(10*np.log(t_max / t_min)))
states = ct.SolutionArray(gas, extra=['t'])
for t in time:
sim.advance(t)
states.append(reactor.thermo.state, t=sim.time)Where reactions:
reactions:
- equation: H3+ + N2 => HN2+ + H2
rate-constant: { A: 1.86e-9, b: 0.0, Ea: 0.0 }
surface-reactions:
- equation: H3+ => H2 + H
sticking-coefficient: { A: 1.0, b: 0.0, Ea: 0.0 }Notice that both reactions require H3+ which is not present in the initial mixture.
Behavior
At some point, molar fraction of "HN2+" becomes negative, and molar fraction of "N2" grows to 1. See the plot
If energy equation is enabled, this also leads to negative temperature.
System information
- Cantera version: 3.1.0 from PyPI
- OS: Ubuntu 25.04
- Python: 3.13.3
Attachments