Linear response functions¶
A general linear response function
can be requested, referring to the linear response of the molecular property associated with Ω^ due to the perturbation associated with V^ and oscillating with the angular frequency ω. The damping term γ is associated with the inverse lifetime of the excited states.
Python script
import veloxchem as vlx
molecule = vlx.Molecule.read_smiles("OO")
basis = vlx.MolecularBasis.read(molecule, "def2-svpd")
scf_drv = vlx.ScfRestrictedDriver()
scf_drv.xcfun = "b3lyp"
scf_results = scf_drv.compute(molecule, basis)
crs = vlx.ComplexResponse()
# available operators
# crs.b_operator = "electric dipole"
# crs.b_operator = "magnetic dipole"
# crs.b_operator = "linear momentum"
# crs.b_operator = "angular momentum"
crs.a_operator = "electric dipole"
crs.b_operator = "magnetic dipole"
crs.a_components = ["x", "y", "z"]
crs.b_components = ["x", "y", "z"]
crs.damping = 0.004556 # 1000 cm-1
crs.frequencies = [0.0656]
crs_results = crs.compute(molecule, basis, scf_results)
Self Consistent Field Driver Setup
====================================
Wave Function Model : Spin-Restricted Kohn-Sham
Initial Guess Model : Superposition of Atomic Densities
Convergence Accelerator : Two Level Direct Inversion of Iterative Subspace
Max. Number of Iterations : 50
Max. Number of Error Vectors : 10
Convergence Threshold : 1.0e-06
ERI Screening Threshold : 1.0e-12
Linear Dependence Threshold : 1.0e-06
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * Starting Reduced Basis SCF calculation...
* Info * ...done. SCF energy in reduced basis set: -150.576224451644 a.u. Time: 0.38 sec.
Iter. | Kohn-Sham Energy | Energy Change | Gradient Norm | Max. Gradient | Density Change
--------------------------------------------------------------------------------------------
1 -151.430265464137 0.0000000000 0.26871108 0.02049182 0.00000000
2 -151.435443369768 -0.0051779056 0.11315252 0.00845747 0.11077370
3 -151.436200333739 -0.0007569640 0.06433828 0.00383362 0.04557585
4 -151.436601355796 -0.0004010221 0.00247619 0.00017252 0.01826297
5 -151.436601924664 -0.0000005689 0.00036657 0.00002378 0.00086748
6 -151.436601936696 -0.0000000120 0.00002855 0.00000242 0.00014612
7 -151.436601936800 -0.0000000001 0.00001093 0.00000054 0.00002014
8 -151.436601936808 -0.0000000000 0.00000618 0.00000046 0.00000386
9 -151.436601936810 -0.0000000000 0.00000162 0.00000012 0.00000171
10 -151.436601936811 -0.0000000000 0.00000004 0.00000000 0.00000049
*** SCF converged in 10 iterations. Time: 2.95 sec.
Spin-Restricted Kohn-Sham:
--------------------------
Total Energy : -151.4366019368 a.u.
Electronic Energy : -190.5210408901 a.u.
Nuclear Repulsion Energy : 39.0844389533 a.u.
------------------------------------
Gradient Norm : 0.0000000441 a.u.
Ground State Information
------------------------
Charge of Molecule : 0.0
Multiplicity (2S+1) : 1
Magnetic Quantum Number (M_S) : 0.0
Complex Response Solver Setup
===============================
Number of Frequencies : 1
Max. Number of Iterations : 150
Convergence Threshold : 1.0e-04
ERI Screening Threshold : 1.0e-12
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * 6 gerade trial vectors in reduced space
* Info * 6 ungerade trial vectors in reduced space
*** Iteration: 1 * Residuals (Max,Min): 1.31e-01 and 1.23e-01
* Info * 12 gerade trial vectors in reduced space
* Info * 11 ungerade trial vectors in reduced space
*** Iteration: 2 * Residuals (Max,Min): 1.66e-02 and 8.19e-03
* Info * 18 gerade trial vectors in reduced space
* Info * 16 ungerade trial vectors in reduced space
*** Iteration: 3 * Residuals (Max,Min): 1.71e-03 and 8.54e-04
* Info * 24 gerade trial vectors in reduced space
* Info * 20 ungerade trial vectors in reduced space
*** Iteration: 4 * Residuals (Max,Min): 3.82e-04 and 1.18e-04
* Info * 28 gerade trial vectors in reduced space
* Info * 25 ungerade trial vectors in reduced space
*** Iteration: 5 * Residuals (Max,Min): 4.31e-05 and 1.05e-05
*** Complex response converged in 5 iterations. Time: 9.45 sec
print(" linear response function")
print(" ------------------------")
print(" frequency real imag")
print("-" * 42)
for key, lrf in crs_results["response_functions"].items():
print(f"{key[0]}{key[1]}{key[2]:12.6f}{lrf.real:16.6f}{lrf.imag:12.6f}")
print("-" * 42) linear response function
------------------------
frequency real imag
------------------------------------------
xx 0.065600 -0.012944 0.172460
yx 0.065600 -0.001355 0.024436
zx 0.065600 -0.184629 2.548857
xy 0.065600 0.013222 -0.185233
yy 0.065600 -0.008363 0.109692
zy 0.065600 -0.018178 0.247708
xz 0.065600 0.120289 -1.685728
yz 0.065600 0.003839 -0.058276
zz 0.065600 0.021634 -0.287929
------------------------------------------
Quadratic response functions¶
A general quadratic response function
can be requested, referring to the linear response of the molecular property associated with Ω^ due to the perturbation associated with V^1 and V^2 oscillating with the angular frequencies ω1 and ω2, respectively. The damping term γ is associated with the inverse lifetime of excited states.
Python script
import veloxchem as vlx
molecule = vlx.Molecule.read_smiles("OO")
basis = vlx.MolecularBasis.read(molecule, "def2-svpd")
scf_drv = vlx.ScfRestrictedDriver()
scf_drv.xcfun = "b3lyp"
scf_results = scf_drv.compute(molecule, basis)
qrf_drv = vlx.QuadraticResponseDriver()
qrf_drv.a_operator = "electric dipole"
qrf_drv.b_operator = "magnetic dipole"
qrf_drv.c_operator = "electric dipole"
# available operators
# qrf.b_operator = "electric dipole"
# qrf.b_operator = "magnetic dipole"
# qrf.b_operator = "linear momentum"
# qrf.b_operator = "angular momentum"
qrf_drv.a_component = "z"
qrf_drv.b_component = "x"
qrf_drv.c_component = "x"
qrf_drv.b_frequencies = [0.0656, 0.1312]
qrf_drv.c_frequencies = [0.0656, 0.1312]
qrf_drv.damping = 0.004556 # 1000 cm-1
qrf_results = qrf_drv.compute(molecule, basis, scf_results)
Self Consistent Field Driver Setup
====================================
Wave Function Model : Spin-Restricted Kohn-Sham
Initial Guess Model : Superposition of Atomic Densities
Convergence Accelerator : Two Level Direct Inversion of Iterative Subspace
Max. Number of Iterations : 50
Max. Number of Error Vectors : 10
Convergence Threshold : 1.0e-06
ERI Screening Threshold : 1.0e-12
Linear Dependence Threshold : 1.0e-06
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * Starting Reduced Basis SCF calculation...
* Info * ...done. SCF energy in reduced basis set: -150.576224459525 a.u. Time: 0.21 sec.
Iter. | Kohn-Sham Energy | Energy Change | Gradient Norm | Max. Gradient | Density Change
--------------------------------------------------------------------------------------------
1 -151.430265937514 0.0000000000 0.26871052 0.02049188 0.00000000
2 -151.435443808597 -0.0051778711 0.11315289 0.00845750 0.11077326
3 -151.436200780757 -0.0007569722 0.06433827 0.00383362 0.04557548
4 -151.436601802524 -0.0004010218 0.00247618 0.00017252 0.01826304
5 -151.436602371384 -0.0000005689 0.00036656 0.00002378 0.00086746
6 -151.436602383417 -0.0000000120 0.00002837 0.00000242 0.00014610
7 -151.436602383522 -0.0000000001 0.00000938 0.00000055 0.00002006
8 -151.436602383530 -0.0000000000 0.00000308 0.00000023 0.00000312
9 -151.436602383530 -0.0000000000 0.00000311 0.00000024 0.00000139
10 -151.436602383530 -0.0000000000 0.00000004 0.00000000 0.00000071
*** SCF converged in 10 iterations. Time: 1.82 sec.
Spin-Restricted Kohn-Sham:
--------------------------
Total Energy : -151.4366023835 a.u.
Electronic Energy : -190.5210506964 a.u.
Nuclear Repulsion Energy : 39.0844483129 a.u.
------------------------------------
Gradient Norm : 0.0000000434 a.u.
Ground State Information
------------------------
Charge of Molecule : 0.0
Multiplicity (2S+1) : 1
Magnetic Quantum Number (M_S) : 0.0
Quadratic Response Driver Setup
=================================
ERI Screening Threshold : 1.0e-12
Convergance Threshold : 1.0e-04
Max. Number of Iterations : 150
Damping Parameter : 4.556000e-03
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
Complex Response Solver Setup
===============================
Number of Frequencies : 1
Max. Number of Iterations : 150
Convergence Threshold : 1.0e-04
ERI Screening Threshold : 1.0e-12
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * 11 gerade trial vectors in reduced space
* Info * 11 ungerade trial vectors in reduced space
*** Iteration: 1 * Residuals (Max,Min): 6.55e-01 and 8.92e-02
* Info * 22 gerade trial vectors in reduced space
* Info * 22 ungerade trial vectors in reduced space
*** Iteration: 2 * Residuals (Max,Min): 8.01e-02 and 9.71e-03
* Info * 33 gerade trial vectors in reduced space
* Info * 33 ungerade trial vectors in reduced space
*** Iteration: 3 * Residuals (Max,Min): 6.85e-03 and 7.76e-04
* Info * 43 gerade trial vectors in reduced space
* Info * 43 ungerade trial vectors in reduced space
*** Iteration: 4 * Residuals (Max,Min): 4.07e-04 and 4.71e-05
* Info * 52 gerade trial vectors in reduced space
* Info * 52 ungerade trial vectors in reduced space
*** Iteration: 5 * Residuals (Max,Min): 4.71e-05 and 6.63e-06
*** Complex response converged in 5 iterations. Time: 52.16 sec
Fock Matrix Computation
=========================
* Info * Processing 4 Fock builds...
* Info * Time spent in Fock matrices: 0.39 sec
Quadratic response function: << z;x,x >> (0.0656,0.0656)
===========================================================
Real Imaginary
----------------------------------------------------
QRF -0.05859285 0.73473106j
Quadratic response function: << z;x,x >> (0.1312,0.1312)
===========================================================
Real Imaginary
----------------------------------------------------
QRF -11.02721386 29.68187647j
*** Time spent in quadratic response calculation: 55.39 sec ***
component = qrf_drv.a_component + qrf_drv.b_component + qrf_drv.c_component
print(" quadratic response function")
print(" ---------------------------")
print(" w1 w2 real imag")
print("-" * 56)
for key, qrf in qrf_results.items():
if key[0] == "qrf":
print(
f"{component} {key[1]:12.6f}{key[2]:12.6f}{qrf.real:16.6f}{qrf.imag:12.6f}"
)
print("-" * 56) quadratic response function
---------------------------
w1 w2 real imag
--------------------------------------------------------
zxx 0.065600 0.065600 -0.058593 0.734731
zxx 0.131200 0.131200 -11.027214 29.681876
--------------------------------------------------------
Cubic response functions¶
A general cubic response function
can be requested, referring to the linear response of the molecular property associated with Ω^ due to the perturbation associated with V^1, V^2, and V^3 oscillating with the angular frequencies ω1, ω2, and ω3, respectively. The damping term γ is associated with the inverse lifetime of excited states.
Python script
import veloxchem as vlx
molecule = vlx.Molecule.read_smiles("OO")
basis = vlx.MolecularBasis.read(molecule, "def2-svpd")
scf_drv = vlx.ScfRestrictedDriver()
scf_drv.xcfun = "b3lyp"
scf_results = scf_drv.compute(molecule, basis)
crf_drv = vlx.CubicResponseDriver()
crf_drv.a_operator = "electric dipole"
crf_drv.b_operator = "magnetic dipole"
crf_drv.c_operator = "electric dipole"
crf_drv.d_operator = "electric dipole"
# available operators
# crf.b_operator = "electric dipole"
# crf.b_operator = "magnetic dipole"
# crf.b_operator = "linear momentum"
# crf.b_operator = "angular momentum"
crf_drv.a_component = "z"
crf_drv.b_component = "x"
crf_drv.c_component = "x"
crf_drv.d_component = "z"
crf_drv.b_frequencies = [0.0656, 0.1312]
crf_drv.c_frequencies = [0.0656, 0.1312]
crf_drv.d_frequencies = [0.0656, 0.1312]
crf_drv.damping = 0.004556 # 1000 cm-1
crf_results = crf_drv.compute(molecule, basis, scf_results)
Self Consistent Field Driver Setup
====================================
Wave Function Model : Spin-Restricted Kohn-Sham
Initial Guess Model : Superposition of Atomic Densities
Convergence Accelerator : Two Level Direct Inversion of Iterative Subspace
Max. Number of Iterations : 50
Max. Number of Error Vectors : 10
Convergence Threshold : 1.0e-06
ERI Screening Threshold : 1.0e-12
Linear Dependence Threshold : 1.0e-06
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * Starting Reduced Basis SCF calculation...
* Info * ...done. SCF energy in reduced basis set: -150.576224481629 a.u. Time: 0.37 sec.
Iter. | Kohn-Sham Energy | Energy Change | Gradient Norm | Max. Gradient | Density Change
--------------------------------------------------------------------------------------------
1 -151.430262992732 0.0000000000 0.26871087 0.02049177 0.00000000
2 -151.435440916156 -0.0051779234 0.11315062 0.00845734 0.11077459
3 -151.436197846254 -0.0007569301 0.06433769 0.00383359 0.04557474
4 -151.436598860558 -0.0004010143 0.00247620 0.00017252 0.01826283
5 -151.436599429428 -0.0000005689 0.00036656 0.00002378 0.00086746
6 -151.436599441459 -0.0000000120 0.00002838 0.00000242 0.00014609
7 -151.436599441565 -0.0000000001 0.00000956 0.00000055 0.00002006
8 -151.436599441572 -0.0000000000 0.00000363 0.00000027 0.00000322
9 -151.436599441572 -0.0000000000 0.00000286 0.00000022 0.00000145
10 -151.436599441573 -0.0000000000 0.00000004 0.00000000 0.00000067
*** SCF converged in 10 iterations. Time: 1.98 sec.
Spin-Restricted Kohn-Sham:
--------------------------
Total Energy : -151.4365994416 a.u.
Electronic Energy : -190.5210554794 a.u.
Nuclear Repulsion Energy : 39.0844560378 a.u.
------------------------------------
Gradient Norm : 0.0000000436 a.u.
Ground State Information
------------------------
Charge of Molecule : 0.0
Multiplicity (2S+1) : 1
Magnetic Quantum Number (M_S) : 0.0
Cubic Response Driver Setup
=============================
ERI Screening Threshold : 1.0e-12
Convergance Threshold : 1.0e-04
Max. Number of Iterations : 150
Damping Parameter : 4.556000e-03
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
Complex Response Solver Setup
===============================
Number of Frequencies : 1
Max. Number of Iterations : 150
Convergence Threshold : 1.0e-04
ERI Screening Threshold : 1.0e-12
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * 12 gerade trial vectors in reduced space
* Info * 12 ungerade trial vectors in reduced space
*** Iteration: 1 * Residuals (Max,Min): 6.46e-01 and 9.57e-02
* Info * 24 gerade trial vectors in reduced space
* Info * 24 ungerade trial vectors in reduced space
*** Iteration: 2 * Residuals (Max,Min): 3.01e-01 and 8.95e-03
* Info * 36 gerade trial vectors in reduced space
* Info * 36 ungerade trial vectors in reduced space
*** Iteration: 3 * Residuals (Max,Min): 1.25e-01 and 8.01e-04
* Info * 48 gerade trial vectors in reduced space
* Info * 48 ungerade trial vectors in reduced space
*** Iteration: 4 * Residuals (Max,Min): 9.94e-02 and 4.84e-05
* Info * 59 gerade trial vectors in reduced space
* Info * 58 ungerade trial vectors in reduced space
*** Iteration: 5 * Residuals (Max,Min): 3.08e-02 and 6.35e-06
* Info * 61 gerade trial vectors in reduced space
* Info * 60 ungerade trial vectors in reduced space
*** Iteration: 6 * Residuals (Max,Min): 9.79e-03 and 6.35e-06
* Info * 63 gerade trial vectors in reduced space
* Info * 62 ungerade trial vectors in reduced space
*** Iteration: 7 * Residuals (Max,Min): 6.61e-03 and 6.35e-06
* Info * 65 gerade trial vectors in reduced space
* Info * 64 ungerade trial vectors in reduced space
*** Iteration: 8 * Residuals (Max,Min): 1.92e-03 and 6.35e-06
* Info * 67 gerade trial vectors in reduced space
* Info * 66 ungerade trial vectors in reduced space
*** Iteration: 9 * Residuals (Max,Min): 3.74e-04 and 6.35e-06
* Info * 69 gerade trial vectors in reduced space
* Info * 68 ungerade trial vectors in reduced space
*** Iteration: 10 * Residuals (Max,Min): 1.97e-04 and 6.35e-06
* Info * 71 gerade trial vectors in reduced space
* Info * 70 ungerade trial vectors in reduced space
*** Iteration: 11 * Residuals (Max,Min): 6.70e-05 and 6.35e-06
*** Complex response converged in 11 iterations. Time: 34.74 sec
Fock Matrix Computation
=========================
* Info * Processing 16 Fock builds...
* Info * Time spent in Fock matrices: 19.47 sec
Complex Response Solver Setup
===============================
Number of Frequencies : 1
Max. Number of Iterations : 150
Convergence Threshold : 1.0e-04
ERI Screening Threshold : 1.0e-12
Exchange-Correlation Functional : B3LYP
Molecular Grid Level : 4
* Info * Using the B3LYP functional.
P. J. Stephens, F. J. Devlin, C. F. Chabalowski, and M. J. Frisch., J. Phys. Chem. 98, 11623 (1994)
* Info * Using the Libxc library (v7.0.0).
S. Lehtola, C. Steigemann, M. J.T. Oliveira, and M. A.L. Marques., SoftwareX 7, 1–5 (2018)
* Info * Using the following algorithm for XC numerical integration.
J. Kussmann, H. Laqua and C. Ochsenfeld, J. Chem. Theory Comput. 2021, 17, 1512-1521
* Info * 12 gerade trial vectors in reduced space
* Info * 12 ungerade trial vectors in reduced space
*** Iteration: 1 * Residuals (Max,Min): 1.74e-01 and 5.67e-02
* Info * 24 gerade trial vectors in reduced space
* Info * 24 ungerade trial vectors in reduced space
*** Iteration: 2 * Residuals (Max,Min): 3.51e-02 and 9.95e-03
* Info * 36 gerade trial vectors in reduced space
* Info * 36 ungerade trial vectors in reduced space
*** Iteration: 3 * Residuals (Max,Min): 6.72e-03 and 6.00e-04
* Info * 48 gerade trial vectors in reduced space
* Info * 48 ungerade trial vectors in reduced space
*** Iteration: 4 * Residuals (Max,Min): 9.48e-04 and 4.34e-05
* Info * 58 gerade trial vectors in reduced space
* Info * 58 ungerade trial vectors in reduced space
*** Iteration: 5 * Residuals (Max,Min): 6.11e-05 and 4.55e-06
*** Complex response converged in 5 iterations. Time: 32.02 sec
Fock Matrix Computation
=========================
* Info * Processing 4 Fock builds...
* Info * Time spent in Fock matrices: 3.77 sec
Cubic response function: << z;x,x,z >> (0.0656,0.0656,0.0656)
================================================================
Real Imaginary
----------------------------------------------------
CRF 2.56720680 -4.22418535j
Cubic response function: << z;x,x,z >> (0.1312,0.1312,0.1312)
================================================================
Real Imaginary
----------------------------------------------------
CRF -2782.58153159 -1021.67617583j
*** Time spent in cubic response calculation: 110.78 sec ***
component = (
crf_drv.a_component
+ crf_drv.b_component
+ crf_drv.c_component
+ crf_drv.d_component
)
print(" cubic response function")
print(" ---------------------------")
print(" w1 w2 w3 real imag")
print("-" * 73)
for key, crf in crf_results.items():
if key[0] == "crf":
print(
f"{component} {key[1]:12.6f}{key[2]:12.6f}{key[3]:12.6f}{crf.real:16.6f}{crf.imag:16.6f}"
)
print("-" * 73) cubic response function
---------------------------
w1 w2 w3 real imag
-------------------------------------------------------------------------
zxxz 0.065600 0.065600 0.065600 2.567207 -4.224185
zxxz 0.131200 0.131200 0.131200 -2782.581532 -1021.676176
-------------------------------------------------------------------------