[Bug]: SolverError when simulating a half-cell and setting initial_soc

[d-cogswell]

PyBaMM Version

25.1.1

Python Version

3.12.4

Describe the bug

working electrode and initial_soc both work independently, but cause a SolverError when used together.

Steps to Reproduce

Here is a working example:

import pybamm

experiment = pybamm.Experiment(["Discharge at 1C until 3V"])
parameter_values = pybamm.ParameterValues("Xu2019")
parameter_values.update(
    pybamm.ParameterValues("Prada2013"), check_already_exists=False
)

# This works
model = pybamm.lithium_ion.SPM()
sim = pybamm.Simulation(model, parameter_values=parameter_values, experiment=experiment)
sol = sim.solve(initial_soc=0.8)

# This works
model = pybamm.lithium_ion.SPM(options={"working electrode": "positive"})
sim = pybamm.Simulation(model, parameter_values=parameter_values, experiment=experiment)
sol = sim.solve()

# This fails
model = pybamm.lithium_ion.SPM(options={"working electrode": "positive"})
sim = pybamm.Simulation(model, parameter_values=parameter_values, experiment=experiment)
sol = sim.solve(initial_soc=0.8)

Relevant log output

pybamm.expression_tree.exceptions.SolverError: Could not find acceptable solution: solver terminated successfully, but maximum solution error (0.00027437641369143506) above tolerance (1e-06)

[d-cogswell]

Switching from scipy.optimize.root() to scipy.optimize.minimize() or scipy.optimize.least_squares() as the algebraic solver seems to fix the issue.

[martinjrobins]

Are you happy with the fix of changing to a different algebraic solver @d-cogswell? If so I'll close this issue.

[agriyakhetarpal]

I assume #4873 is supposed to close this, but it's a draft and isn't linked to this issue yet.