标题: 求助：原子丢失 [打印本页] 作者Author: kidult 时间: yesterday 22:48 标题: 求助：原子丢失 体系密度0.1，约3000个原子，在nvt系综300k下弛豫10ps，然后在nvt、1000k下弛豫10ps，接着把温度提升到3000k恒温燃烧。
出现”原子丢失“报错。试过很多方法（neighbor增大，减小时间步长）都不行
下面是我的in文件
==================== 基本模拟设置 ====================
units real
atom_style charge
boundary p p p
read_data C16H34.data
neighbor 3 bin
neigh_modify every 1 delay 0 check no
timestep 0.1
# ==================== 力场设置 ====================
pair_style reaxff NULL
pair_coeff * * ffield.reax.cho C C H C C C
fix 02 all qeq/reax 1 0.0 10.0 1e-6 reaxff
# ==================== 能量最小化阶段 ====================
thermo 100
thermo_style custom step press pe temp ke
dump 1 all atom 100 minimize.xyz
minimize 1e-6 1e-6 5000 10000
undump 1
reset_timestep 0
# ==================== 第一阶段：300K NVT弛豫 (10ps) ====================
thermo 1000
thermo_style custom step press pe temp ke
dump 2 all atom 1000 relax_300K.xyz # 修改输出文件名
fix nvt all nvt temp 300 300 100
fix 2 all temp/berendsen 300 300 100
run 10000 # 100ps
unfix nvt
undump 2
reset_timestep 0
# ==================== 第一阶段：1000K NVT弛豫 (10ps) ====================
thermo 1000
thermo_style custom step press pe temp ke
dump 2 all atom 1000 relax_300K.xyz # 修改输出文件名
velocity all create 1500 4928459
fix nvt all nvt temp 1000 1000 100
fix 2 all temp/berendsen 1000 1000 100
run 100000 # 10ps
unfix nvt
undump 2
reset_timestep 0
# ==================== 燃烧模拟阶段 ====================
velocity all create 3000.0 4928459 rot yes dist gaussian
fix nvt all nvt temp 3000 3000 50
fix 3 all temp/berendsen 3000 3000 50
fix species all reaxff/species 50 100 5000 species2.log element C C H C C C
fix reaxff_bonds all reaxff/bonds 10000 bonds.log
dump 4 all atom 10000 combustion.xyz
thermo 1000
thermo_style custom step temp press pe ke density
run 30000000