求助：高斯ts计算结果虚频正确，报错link9999，结果可信嘛？

Bayanbuluk

我在计算一个自由基反应过程R-CF2COO·→R-CF2· + CO2的过程，用 #p opt=(calcall,ts) freq ub3lyp/6-31g(d) scrf=(iefpcm,solvent=water) em=gd3bj计算的过渡态，计算的结果显示虚频只有一个，振动方向看起来也是CO2逃离整个分子的方向，但是gaussian的log文件没有显示正常结束，反而报错link 9999，我向上翻到了这样的一个warning： Principal axes and moments of inertia in atomic units:
                           1         2         3
     Eigenvalues --  ******************************
           X            0.99999   0.00164   0.00307
           Y           -0.00167   0.99995   0.01010
           Z           -0.00306  -0.01011   0.99994
This molecule is an asymmetric top.
Rotational symmetry number  1.
Warning -- assumption of classical behavior for rotation
           may cause significant error
Rotational temperatures (Kelvin)      0.00730     0.00187     0.00171
Rotational constants (GHZ):           0.15206     0.03896     0.03570
    1 imaginary frequencies ignored.
Zero-point vibrational energy     462514.8 (Joules/Mol)
                                  110.54370 (Kcal/Mol)
Warning -- explicit consideration of  72 degrees of freedom as
           vibrations may cause significant error

log文件结尾是这样的
HE WHO LOVES TO READ, AND KNOWS HOW TO REFLECT,
HAS LAID BY A PERPETUAL FEAST FOR HIS OLD AGE.
-- UNCLE ESEK, "SCRIBNER'S MONTHLY", SEPT. 1880
Error termination request processed by link 9999.
Error termination via Lnk1e in /data/home/xclyangweiguang//g09/l9999.exe at Thu May 13 19:29:12 2021.
Job cpu time:       0 days 11 hours 11 minutes 23.6 seconds.
File lengths (MBytes):  RWF=   4988 Int=      0 D2E=      0 Chk=     52 Scr=      4

请问一下各位前辈，这个报错对ts计算结果有影响嘛？是否可以拿这个结构来进行接下来的irc计算。如果不行，该怎么改进呢？我添加了log和fchk文件在下面

sobereva

有报错
Optimization stopped.
    -- Wrong number of Negative eigenvalues: Desired=  1 Actual=  2

找过渡态始终加上noeigen