卡开发发 发表于 2025-6-12 03:54
1、
这个问题应该是三个轴向k网格差距过大导致的警告，但并不妨碍继续计算。但你的结构是立方结构，a=b=c ...

好的，谢谢你的解答。

1、

VERY BAD NEWS! internal error in subroutine IBZKPT:Reciprocal lattice and k-lattice belong to different class of lattices. Often results are still useful.

这个问题应该是三个轴向k网格差距过大导致的警告，但并不妨碍继续计算。但你的结构是立方结构，a=b=c，似乎没有道理k点取3×3×2。
2、变晶胞优化尽可能保证计算精度充分，且需要拷贝结构多优化几次（这是原理上所要求的，但不是唯一方案）。
3、文献未提到PREC和赝势具体设置的信息，且从提供的文献信息看似乎vdW修正的方式似乎也不是特别明确，这都有可能造成差异。

北大-陶豫 发表于 2025-6-11 11:36
我还是建议你直接把文件作为附件传上来吧，太大的文件可以压缩一下

好的，已经把文件上传了。

我还是建议你直接把文件作为附件传上来吧，太大的文件可以压缩一下

To describe the interactions between core and valence electrons, the pro­ jector augmented wave (PAW) method was employed.
For hybrid perovskite materials, incorporating van der Waals interaction correc­ tions typically enhances the accuracy of the results.
In the aspect of exchange correlation function selection, this study selects the Per­dew, Burke, and Ernzerh of (PBE) functional.
A plane-wave cutoff energy of 450 eV was set, and the conjugate gradient algorithm was utilized for atomic position optimi­ zation.
The convergence criteria for the Hellmann-Feynman force and energy were established at 0.03 eV/Å and 1 × 10–5 eV, respectively.
The Gammacentered Monkhorst-Pack k-point mesh employed were set as 3 × 3 × 2 in geometry optimizations.