Complex Scaling on Q-Chem

Questions
#1

Hello,
I want to calculate the Mg- resonance using the complex scaling eom-ea-ccsd method, but the calculation refuses to run.
More specifically, Q-Chem only performs the coupled-cluster calculation but completely ignores the eom-ea-ccsd part and the complex scaling part.
When I perform the eom-ea-ccsd calculation without complex scaling, Q-Chem produces the complete response.
I admit that I do not know how to solve this problem, as the qchem manual does not detail complex scaling.

Here is my input:
$comment
Mg- shape resonance using CS-EOM-EA-CCSD
$end

$molecule
0 1
Mg 0.0 0.0 0.0
$end

$rem
METHOD EOM-CCSD
BASIS gen
EA_STATES [2,2,2,2,2,2,2,2]
CC_EOM_PROP true
COMPLEX_CCMAN 1
CC_MEMORY 4000
MEM_TOTAL 8000
$end

$complex_ccman
CS_HF true
CS_CCSD true
CS_THETA 10
$end

$basis
Mg 0
S 15 1.00
1.649000D+05 7.299290D-05
2.471000D+04 5.666520D-04
5.628000D+03 2.962690D-03
1.596000D+03 1.229620D-02
5.210000D+02 4.273240D-02
1.880000D+02 1.230130D-01
7.301000D+01 2.748320D-01
2.990000D+01 4.018180D-01
1.254000D+01 2.646970D-01
4.306000D+00 3.326120D-02
1.826000D+00 -4.413350D-03
7.417000D-01 2.060240D-03
1.457000D-01 -8.384700D-04
7.612000D-02 7.081950D-04
3.310000D-02 -2.006020D-04
S 15 1.00
1.649000D+05 -1.842480D-05
2.471000D+04 -1.435000D-04
5.628000D+03 -7.487100D-04
1.596000D+03 -3.144070D-03
5.210000D+02 -1.104810D-02
1.880000D+02 -3.360580D-02
7.301000D+01 -8.259460D-02
2.990000D+01 -1.593140D-01
1.254000D+01 -1.528880D-01
4.306000D+00 1.908490D-01
1.826000D+00 5.799640D-01
7.417000D-01 3.720290D-01
1.457000D-01 2.116910D-02
7.612000D-02 -1.193440D-02
3.310000D-02 2.773290D-03
S 1 1.00
1.457000D-01 1.000000D+00
S 15 1.00
1.649000D+05 3.551760D-06
2.471000D+04 2.764200D-05
5.628000D+03 1.444040D-04
1.596000D+03 6.057440D-04
5.210000D+02 2.135270D-03
1.880000D+02 6.499340D-03
7.301000D+01 1.614460D-02
2.990000D+01 3.157660D-02
1.254000D+01 3.163740D-02
4.306000D+00 -4.391400D-02
1.826000D+00 -1.510930D-01
7.417000D-01 -2.176680D-01
1.457000D-01 2.598070D-01
7.612000D-02 5.472450D-01
3.310000D-02 3.259100D-01
S 1 1.00
3.310000D-02 1.000000D+00
S 1 1.00
1.200000D-02 1.000000D+00
S 1 1.00
6.000000D-03 1.000000D+00
P 10 1.00
3.169000D+02 2.075320D-03
7.486000D+01 1.628690D-02
2.372000D+01 7.386970D-02
8.669000D+00 2.142970D-01
3.363000D+00 3.821540D-01
1.310000D+00 3.981780D-01
4.911000D-01 1.528780D-01
2.364000D-01 -4.375400D-03
8.733000D-02 6.295820D-03
3.237000D-02 -1.567690D-03
P 1 1.00
8.733000D-02 1.000000D+00
P 10 1.00
3.169000D+02 -3.297270D-04
7.486000D+01 -2.587540D-03
2.372000D+01 -1.191200D-02
8.669000D+00 -3.502270D-02
3.363000D+00 -6.399680D-02
1.310000D+00 -7.044360D-02
4.911000D-01 -3.758360D-02
2.364000D-01 1.770430D-01
8.733000D-02 5.667070D-01
3.237000D-02 3.961430D-01
P 1 1.00
3.237000D-02 1.000000D+00
P 1 1.00
9.000000D-03 1.000000D+00
P 1 1.00
4.400000D-03 1.000000D+00
D 1 1.00
2.998000D-01 1.000000D+00
D 1 1.00
1.285000D-01 1.000000D+00
D 1 1.00
4.440000D-02 1.000000D+00
D 1 1.00
2.200000D-02 1.000000D+00
F 1 1.00
2.417000D-01 1.000000D+00
F 1 1.00
8.650000D-02 1.000000D+00
F 1 1.00
4.325000D-02 1.000000D+00

$end

Thank you in advance.

Translated with DeepL.com (free version)

#2

Can you please give more details how the calculation progresses? Ideally, it would be most helpful to the output. You can upload file to a google drive and post a link in your response.

#3

As a scientific recommendation, straight CS is not very good – CAP or CBF calculations are much more robust.

Here is what you need to use to run CAP-EOM-EA:

$complex_ccman
CS_HF 1
CAP_TYPE 1
CAP_ETA 10000
CAP_X 2000
CAP_Y 2500
CAP_Z 2500
$end

And for CBF:

$complex_ccman
cs_theta = 0
cs_alpha = 100
$end

There are examples for such jobs in the $QC/samples/ccman2 directory.

#4

I figured out – your input will work with this modification:

$complex_ccman
CS_HF 1
CS_CCSD 1
CS_THETA 10
$end

Within complex_ccman section, you cannot substitute integers by boolean.

My advice about CS vs CBF or CAP still stands.