Rep:Mod:MH5015-2
Y2 Inorganic Computational Chemistry lab report by Martin Holicky.
Borane
The structure of a borane molecule was optimised and its vibrational frequencies analysed.
Structure optimisation
Method: B3LYP/6-31g(d,p)
Summary table:
File Name MH_BH3_SYM_OPT File Type .log Calculation Type FOPT Calculation Method RB3LYP Basis Set 6-31G(d,p) Charge 0 Spin Singlet E(RB3LYP) -26.61532342 a.u. RMS Gradient Norm 0.00009608 a.u. Dipole Moment 0.0000 Debye Point Group D3H
Item table:
Item Value Threshold Converged? Maximum Force 0.000192 0.000450 YES RMS Force 0.000126 0.000300 YES Maximum Displacement 0.000762 0.001800 YES RMS Displacement 0.000499 0.001200 YES
Output log: File:MH BH3 SYM OPT.LOG
Optimised Borane molecule |
Molecular orbitals
The computed MOs match the MOs from the qualitative theory very well. Of course, the qualitative theory can predict the lobe shapes and sizes to only some extent but it is still quite close to the real orbitals. Probably the orbital most different from the theoretical was the antibonding a1', with the central lobe being much smaller than drawn. The diagram was based on [1].
Frequency analysis
Using the structure computed in the previous step, the infrared vibrations were analysed.
Method: B3LYP/6-31g(d,p)
Summary table:
File Name MH_BH3_freq File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set 6-31G(d,p) Charge 0 Spin Singlet E(RB3LYP) -26.61532342 a.u. RMS Gradient Norm 0.00009609 a.u. Dipole Moment 0.0000 Debye Point Group D3H
Item table:
Item Value Threshold Converged? Maximum Force 0.000192 0.000450 YES RMS Force 0.000096 0.000300 YES Maximum Displacement 0.000757 0.001800 YES RMS Displacement 0.000378 0.001200 YES
Output log: File:MH BH3 FREQ.LOG
IR Frequencies:
Low frequencies --- -0.1082 -0.0046 0.0007 46.3499 46.3501 47.3456 Low frequencies --- 1163.7050 1213.6299 1213.6302
The zero frequencies are higher than the +-15 cm-1 ideal, but the demonstrators have suggested to proceed with the analysis anyway - the zero frequencies could not be decreased even with the grid=ultrafine setting. The main frequencies (1163, 1213 cm-1) are exactly the same as in the lab manual.
Infrared spectrum
The following IR spectrum was computed:
| wavenumbers (cm-1) | IR active? | symmetry | Intensity (arb. u.) | vibration type |
| 1164 | yes | A2' ' | 92 | out-of-plane bend |
| 1213 | yes | E' | 14 | in-plane bend |
| 1213 | yes | E' | 14 | in-plane bend |
| 2580 | no | A1' | 0 | symmetric stretch |
| 2713 | yes | A2' ' | 126 | asymmetric stretch |
| 2713 | yes | A2' ' | 126 | asymmetric stretch |
There are 3N-6 = 6 vibrational modes. There are two pairs of degenerate modes (1213 and 2713 cm-1) and an IR inactive mode due to symmetry (2580 cm-1). This leaves the three observable peaks.
Borane-ammonia reaction
In order to calculate the dissociation energy of BH3NH3 (the reaction below), the RB3LYP energies were computed for each of the participating compounds.
BH3NH3 → BH3 + NH3
Ammonia
Summary table:
File Name MH_NH3_OPT File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set 6-31G(d,p) Charge 0 Spin Singlet E(RB3LYP) -56.55776873 a.u. RMS Gradient Norm 0.00000323 a.u. Imaginary Freq 0 Dipole Moment 1.8465 Debye Point Group C3V
Item table:
Maximum Force 0.000006 0.000450 YES RMS Force 0.000004 0.000300 YES Maximum Displacement 0.000014 0.001800 YES RMS Displacement 0.000009 0.001200 YES
Output log: File:MH NH3 OPT.LOG
Borane-ammonia adduct
Summary table:
File Name MH_BH3NH3_OPT File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set 6-31G(d,p) Charge 0 Spin Singlet E(RB3LYP) -83.22468893 a.u. RMS Gradient Norm 0.00005975 a.u. Imaginary Freq 0 Dipole Moment 5.5651 Debye Point Group C1
Item table:
Item Value Threshold Converged? Maximum Force 0.000122 0.000450 YES RMS Force 0.000058 0.000300 YES Maximum Displacement 0.000582 0.001800 YES RMS Displacement 0.000320 0.001200 YES
Output log: File:MH BH3NH3 OPT.LOG
The reaction
The energy difference between the reactant and the products, using the calculations above, the following:
E(NH3) = -56.55776 a.u.
E(BH3) = -26.61532 a.u.
E(NH3BH3) = -83.22468 a.u.
ΔE=E(NH3BH3)-[E(NH3)+E(BH3)] = -136 kJ/mol (-0.05160 a.u.)
The value of the dissociation energy calculated appears to be reasonable and in the expected range.
Smf115 (talk) 22:31, 15 May 2018 (BST)Clear calculation and correct accuracy for the final energy values. To improve a discussion of the bond strength by comparison to other, referenced, bond dissociation energies would be more relevant.
Boron tribromide
In order to optimise the structure of BBr3, pseudopotentials (approximating the core electrons as a constant) were used to simplify the calculations.
Method: B3LYP; 6-31g(d,p) (Boron), LanL2DZ (Bromines)
Summary table:
File Name MH_BBr3_FREQ File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set Gen Charge 0 Spin Singlet E(RB3LYP) -64.43644947 a.u. RMS Gradient Norm 0.00000384 a.u. Imaginary Freq 0 Dipole Moment 0.0000 Debye Point Group D3H
Item table:
Item Value Threshold Converged? Maximum Force 0.000008 0.000450 YES RMS Force 0.000005 0.000300 YES Maximum Displacement 0.000036 0.001800 YES RMS Displacement 0.000024 0.001200 YES
IR Frequencies:
Low frequencies --- -2.3080 -0.0030 -0.0018 0.0773 0.7455 0.7455 Low frequencies --- 155.9402 155.9405 267.6897
Simulation output: DOI:10042/202301
Optimised BBr3 molecule |
Simulation of the trans influence in platinum (II) complexes
There are multiple types of bonding involved in the formation of complexes and pi-bonding, depending on the ligand, also contributes. This contribution not only affects the ligand itself but also the ligand trans to it. This is called the trans influence and in this report it will be studied on two square-planar Pt(II) complexes, one with a strong trans-effect ligand (CO) and one with a weak trans-effect ligand (NH3). Additionally, a symmetric complex with no trans influence (PtCl4-2) will be investigated in order to compare its computed properties to the well-known literature values.
Calculations
The full basis set (6-31g(d,p)) was used for all atoms except platinum, which was approximated using LanL2DZ pseudopotentials.
Phunt (talk) 05:59, 17 May 2018 (BST)you should also include the method employed here as well
PtCl4-2
Method: B3LYP; 6-31g(d,p) (Chlorine), LanL2DZ (Platinum)
Summary table:
File Name MH_PTCL4 File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set Gen Charge -2 Spin Singlet E(RB3LYP) -1960.11792527 a.u. RMS Gradient Norm 0.00000040 a.u. Imaginary Freq 0 Dipole Moment 0.0000 Debye Point Group D4H
Item table:
Item Value Threshold Converged? Maximum Force 0.000001 0.000450 YES RMS Force 0.000000 0.000300 YES Maximum Displacement 0.000008 0.001800 YES RMS Displacement 0.000004 0.001200 YES
Low Frequencies:
Low frequencies --- -3.8310 -0.0041 -0.0034 -0.0028 5.0476 5.0476 Low frequencies --- 74.4791 124.6450 138.4069
Output log: File:MH PTCL4.LOG
Phunt (talk) 06:01, 17 May 2018 (BST)please link to the files and not to the page that links to the file
PtCl3(NH3)-1
Method: B3LYP; 6-31g(d,p) (Chlorines, Nitrogen, Hydrogens), LanL2DZ (Platinum)
Summary table:
File Name MH_PTCL3NH3 File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set Gen Charge -1 Spin Singlet E(RB3LYP) -1556.52602330 a.u. RMS Gradient Norm 0.00008959 a.u. Imaginary Freq 0 Dipole Moment 7.8905 Debye Point Group C1
Item table:
Item Value Threshold Converged? Maximum Force 0.000177 0.000450 YES RMS Force 0.000067 0.000300 YES Maximum Displacement 0.001676 0.001800 YES RMS Displacement 0.000602 0.001200 YES
Low Frequencies:
Low frequencies --- -2.1104 0.0021 0.0025 0.0028 1.2331 5.0423 Low frequencies --- 35.8683 102.6610 126.8053
Output log: File:MH PTCL3NH3.LOG
Phunt (talk) 06:03, 17 May 2018 (BST)a comment that your frequency file does not show the same level of convergence would have been even better
Item Value Threshold Converged? Maximum Force 0.000177 0.000450 YES RMS Force 0.000067 0.000300 YES Maximum Displacement 0.005019 0.001800 NO RMS Displacement 0.002113 0.001200 NO
PtCl3(CO)-1
Method: B3LYP; 6-31g(d,p) (Chlorine, Carbon, Oxygen), LanL2DZ (Platinum)
Summary table:
File Name MH_PTCL3CO File Type .log Calculation Type FREQ Calculation Method RB3LYP Basis Set Gen Charge -1 Spin Singlet E(RB3LYP) -1613.29736890 a.u. RMS Gradient Norm 0.00001757 a.u. Imaginary Freq 0 Dipole Moment 3.9566 Debye Point Group C2V
Item table:
Item Value Threshold Converged? Maximum Force 0.000045 0.000450 YES RMS Force 0.000020 0.000300 YES Maximum Displacement 0.000525 0.001800 YES RMS Displacement 0.000230 0.001200 YES
Low Frequencies:
Low frequencies --- -0.0041 -0.0036 -0.0014 3.1187 3.9560 5.9442 Low frequencies --- 72.7598 92.4198 111.4315
Output log: File:MH PTCL3CO.LOG
Infrared spectra
The complexes had a large number of vibrational modes. Therefore, the two most interesting modes are reported here. For all complexes they were strongly IR-active.
| Complex | Pt-Cl stretch (trans to Cl) (cm-1) | Pt-Cl stretch (trans to L) (cm-1) |
| [PtCl4]2- | 267 | 267 |
| [PtCl3NH3]-1 | 291 | 318 |
| [PtCl3CO]-1 | 309 | 312 |
With increasing back-bonding the bond strength increases and therefore also the infrared frequencies. Comparing the Pt-Cl stretch frequencies (trans to L), the Pt-Cl bond trans to NH3 is stronger than the bond trans to CO, indicating the CO group displays a stronger trans-effect. This is in an agreement with the expected behaviour - the CO ligand is a good pi-acceptor (while the ammonia is only a weak sigma-donor) and therefore causes a stronger trans effect [2]. However, a more significant change appeared in the chlorines cis to the ligand - this might be related to the different charges of the complexes and also a slightly bent geometry (see the next section), affecting the stretch vectors.
The literature value for the stretches of [PtCl4]2- is 320 cm-1 [3]. The calculated value of 267 cm-1 therefore presents a 20% error. A more accurate and detailed basis set could improve the accuracy of the simulation.
Interestingly, the C-O triple bond stretch appears as a strong peak at 2136 cm-1. The position of this peak is diagnostic for pi back-bonding (lower frequencies as a result of weaker C-O bond because of the donation into the antibonding orbitals) [4]. However, the current peak frequency corresponds to free carbon monoxide (2143 cm-1), even though it should be significantly lower - the accuracy of the computation is, of course, limited.
Phunt (talk) 06:10, 17 May 2018 (BST)very nice attempt at analysis, you are asking all the right questions. You don't quite yet have all the knowledge and experience yet to make full answers. While one reason for the miss-match with the C-O vibration may be the computational method, however it could also mean the current theory of back-bonding is over-simplified, a common problem when you really try to understand the bonding in many cases. The most interesting areas of research are where just such discrepancies occur!
Structures
| Complex | Pt-Cl bond (trans to L) (Å) | Pt-Cl bond (cis to L) (Å) | Cl-Pt-Cl angle (degrees) |
| [PtCl4]2- | 2.42 | 2.42 | 180 |
| [PtCl3NH3]-1 | 2.36 | 2.41 | 170.8 |
| [PtCl3CO]-1 | 2.37 | 2.41 | 178.2 |
Two interesting phenomena were observed on the structures - the Cl-Pt-Cl bond angle deviated from the ideal 180 degrees for steric reasons, especially for the complex with ammonia. Also, the Pt-Cl bond trans to CO/NH3 was longer for the CO complex. This is expected due to the strong trans influence of the CO ligand weakening the Pt-Cl bond and thus making it longer.
The [PtCl4]2- bond length (2.42 Å) approximately corresponds to the literature (crystallographic) value of 2.30 Å[5], though the deviation is significant.
Optimised PtCl4-2 molecule |
Optimised PtCl3NH3 molecule |
Optimised PtCl3CO molecule |
Molecular orbitals
The point groups of the molecules were D4h ([PtCl4]2-), C2v ([PtCl3CO]-1]) and C1 ([PtCl3NH3]-1). The decreasing symmetry makes it harder it use qualitative MO theory - especially in the case of the ammonia complex which has no symmetry at all. In the case of the carbonyl complex, there are 4 different symmetry labels (as opposed to 10 in D4h). This, again, complicates the diagrams since mixing is expected between the orbitals with the same symmetry, splitting the MOs and increasing the number of different energy levels. Therefore, the orbitals for the least complicated case, [PtCl4]2-, were analysed.
Phunt (talk) 06:13, 17 May 2018 (BST) the reduction of symmetry does not generally have a large effect on the MO makeup, and normally we run all calculations in C1 to ensure we have the optimised geometry, so the above does not really apply. You know how to assign the symmetry labels even if the code has not done it for you.
Three chosen MOs are in the figure below.
Phunt (talk) 06:19, 17 May 2018 (BST) some nice orbitals chosen, especially MO57 which is showing very strong through space interactions, not were we normally think of the bonds. MO65 careful, we don't call antibonding interactions "overlap" as there is a node the overlap is minimal, but there is still an "interaction". The orbitals involved are all pAOs not dAOs (just a slip up?) Really nice to identify the radial node, well done!
An interesting point is that most of the orbitals should have one or mode radial nodes (e.g. 2 for the platinum 5d). However, the observed number of radial nodes is lower than expected - this is most likely due to the basis/pseudopotential set LanL2DZ omitting the inner nodes to simplify the simulations.
In the MOs of [PtCl3CO]-1 it was possible to clearly observe the pi-donation from the platinum d-orbital to the carbon monoxide π* orbital, as shown in the render below (MO 32, filled).
Phunt (talk) 06:19, 17 May 2018 (BST) correct on both counts, very nice!
Charges
The Mulliken charges on the platinum and chlorine atoms are summarised below -
Phunt (talk) 06:25, 17 May 2018 (BST)Mulliken charges are unreliable, you should report NBO charges as covered in the main part of the lab. Nevertheless the conclusions you draw based on these are correct.
| Complex | Cl (trans to L) charge | Cl (cis to L) charge | Pt charge |
| [PtCl4]2- | -0.44 | -0.44 | -0.26 |
| [PtCl3NH3]-1 | -0.32 | -0.39 | -0.15 |
| [PtCl3CO]-1 | -0.31 | -0.29 | -0.01 |
The pi-accepting abilities of the carbonyl group are clearly visible - the overall charges on Pt and Cl are smaller (less electron density) compared to the other two complexes. However, the comparison between [PtCl4]2- and the other complexes is limited, because the complex has a higher net charge. It would be certainly interesting to simulate the platinum complexes with other ligands with varying levels of sigma/pi donation/acceptance abilities for a better comparison.
Phunt (talk) 06:25, 17 May 2018 (BST)Yes this has been done you might like to check Chen, Y; Hartmann, M; Frenking, G, ZEITSCHRIFT FUR ANORGANISCHE UND ALLGEMEINE CHEMIE Volume: 627 Issue: 5 Pages: 985-998 Published: MAY 2001
Conclusions
Using Gaussian molecular simulations, it was possible to observe the trans influence on the bond lengths and vibrational frequencies in square-planar Pt(II) complexes. However, the accuracy of the simulations was limited and it would be both interesting and statistically meaningful to perform the comparisons with more than two complexes.
Phunt (talk) 06:28, 17 May 2018 (BST)Overall an excellent report, a couple of conceptual points that I hope are clearer for you now.
- ↑ Hunt, T.: 2nd Year Molecular Orbitals lecture notes. 2017, Imperial College London.
- ↑ Atkins, P. W. et al. Shriver and Atkins' Inorganic Chemistry, 5ed. 2010, Oxford University Press.
- ↑ Wild U., et al. Eur. J. Inorg. Chem. 2008, 1248–1257.
- ↑ Long, N. 2nd year Transition Metal lecture notes. 2018, Imperial College London.
- ↑ Elmali A., et al. Z. Naturforch, B: Chem. Sci. 2015, 60, 164.