I have been trying to optimize the geometry of a copper based metal halide perovskite. I opened the cif file using BURAI GUI and generated the vc-relax input file. I have increased the steps to be 100 but it's not converging either. should I increase it more or making converging not necessary or change the geometry? I'm using 222 geometry and the structure is basically 0-D. I am attaching my input file and a photo of the structure
Try this:
When performing geometry optimization, it's important to consider several factors:
1. Convergence criteria: The convergence criteria determine when the optimization is considered converged. This can be based on energy changes, forces, or other parameters. Check your optimization input file to ensure that the convergence criteria are appropriately set. You may need to adjust these criteria based on the specific characteristics of your system.
2. Number of steps: Increasing the number of optimization steps can potentially help improve convergence. However, there may be other factors affecting convergence, such as the initial geometry, choice of optimization algorithm, or potential energy surface landscape. It's worth experimenting with a higher number of steps to see if it improves convergence, but keep in mind that excessively increasing the number of steps may not necessarily lead to convergence if there are other underlying issues.
3. Initial geometry: The initial structure you provide for the geometry optimization can greatly influence the convergence. Ensure that your initial structure is reasonable and close to the expected optimized structure. If the initial geometry is too far from the optimized structure, it may take longer for the optimization to converge, or it may not converge at all.
4. Choice of optimization method: Different optimization algorithms and methods may have varying levels of success depending on the system and its complexity. It's worth exploring different optimization methods to see if they improve convergence. Common methods include conjugate gradient, quasi-Newton, or more advanced approaches like density functional theory (DFT) or molecular dynamics (MD).
5. System size: The size and complexity of the system can impact the convergence of the optimization. Larger and more complex systems may require more iterations to converge. Additionally, if the system contains flexible or dynamic components, it may be more challenging to achieve convergence.
Good luck
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability