View source for Rep:Module2:aiwobiezou

=Preliminary Questions=

{| class="wikitable" border="1"
|-
! No.
! Motion
! Frequency/cm<sup>-1</sup>
! Intensity
! Symmetry - D<sub>3h</sub>
|-
| 1 
| [[Image:Chi-Vib1.jpg|thumb|100px| ]]
| 1146 
| 93 
| A''<sub>2</sub>
Wag 
|-
| 2 
| [[Image:Chi-Vib2.jpg|thumb|100px| ]] 
| 1205 
| 12 
| E'
Scissor
|-
| 3 
| [[Image:Chi-Vib3.jpg|thumb|100px| ]] 
| 1205 
| 12 
| E' 
Rock
|-
| 4 
| [[Image:Chi-Vib4.jpg|thumb|100px| ]]  
| 2592 
| 0 
| A'<sub>1</sub>  
Stretch - Symmetric
|-
| 5 
| [[Image:Chi-Vib5.jpg|thumb|100px| ]] 
| 2730 
| 104 
| E' 
Stretch - Assymetric
|-
| 6 
| [[Image:Chi-Vib6.jpg|thumb|100px| ]] 
| 2730 
| 104 
| E' 
Stretch - Symmetric and Asymmetric
|-
|}

[[Image:Chi-BH3_IR.jpg|thumb|600px|left|IR Spectrum of BH<sub>3</sub> ]]

<br clear="all" />

=Structural Analysis of BCl<sub>3</sub>=

<jmol>
<jmolAppletButton>
<uploadedFileContents>Chi-BCl3.mol</uploadedFileContents>
<text>Jmol</text>
</jmolAppletButton>
</jmol> of BCl<sub>3</sub>

The D<sub>3h</sub> ground state structure of BCl<sub>3</sub> was optimised using the B3YLP method the 3-21G basis set. The calculation took 18.0 seconds. The prodecure followed for [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/2a_optimising_bh3.html optimisation] can be found by clicking on the link.

Optimisation terminates once the gradient of the energy of the molecule reaches (almost) zero, however both a maximum (the transition state of a dissociation energy barrier) and minimum (the ground state) have a gradient of zero. Frequency analysis is important because it is the second derivative of the potential energy surface and hence gives the curvature of the function. If the calculated frequencies are positive we have a minimum and if the frequencies are negative we have a maximum. We can use this information to see whether the molecule has been optimised completely. Frequency analysis also provides the IR and Raman modes.

The same method and basis set must be used for both optimisation and frequency calculations becauase the same accuracy in the description of the molecular orbitals is required. If not it can lead to inaccurate results - i.e. when computing the molecular orbitals using different basis sets would lead to "excited" electron density.

In some structure GaussView does not draw in the bonds where we expect, however this does not mean there is no bond. GaussView draws bonds based on a distance critera so if the distance exceeds some pre-defined value the bond will not be present. Nevertheless this is not important because bonds are a structural convenience. What is important is the prescence of electrons and its intereactions with the the two nuclei.

A bond can be defined as an interaction between the nuclei and electrons of two atoms. Having opposite charges the nuclei and electrons attract each other via electromagnetic force. Bonds are often thought of as electrons between two nuclei because it is the most stable configuration and hence where electrons spend most of their time.

[http://hdl.handle.net/10042/to-3981 Output Files] for b3lyp/3-21g BCl<sub>3</sub> optimisation

[http://hdl.handle.net/10042/to-3980 Output Files] for b3lyp/3-21g BCl<sub>3</sub> frequency

{| class="wikitable" border="1"
! 
! B-Cl Bond Distance
/Angstroms
! Cl-B-Cl Angle
/Degrees
|-
| Calculated || 1.78 || 120
|-
| Literature<ref>F. Bessac, G. Frenking, Inorg. Chem. 2003, 42, 7990-7994. [http://dx.doi.org/10.1021/ic034141o DOI: 10.1021/ic034141o]</ref> || 1.75 || 120
|}
<br clear="all" />
The calculated Cl-B-Cl angle is what is expected. BCl<sub>3</sub> is a simple, trigonal planar molecule where 120 degrees Cl-B-Cl angles would allow the three chlorine atoms to be furthest apart from each other. The calculated B-Cl bond distance, on the other hand, differs from the literature value. This is proabably due to the low accuracy of the basis set used combined with its error(s).

{| class="wikitable" border="1"
|-
! No.
! Motion
! Frequency/cm<sup>-1</sup>
! Intensity
! Symmetry - D<sub>3h</sub>
|-
| 1 
| [[Image:Chi-Vib1.jpg|thumb|100px| ]]
| 260
| 2
| A''<sub>2</sub>
Wag 
|-
| 2 
| [[Image:Chi-Vib2.jpg|thumb|100px| ]] 
| 260 
| 2
| E'
Scissor
|-
| 3 
| [[Image:Chi-Vib3.jpg|thumb|100px| ]] 
| 440 
| 0
| E' 
Rock
|-
| 4 
| [[Image:Chi-Vib4.jpg|thumb|100px| ]]  
| 456
| 11
| A'<sub>1</sub>  
Stretch - Symmetric
|-
| 5 
| [[Image:Chi-Vib5.jpg|thumb|100px| ]] 
| 886 
| 341
| E' 
Stretch - Assymetric
|-
| 6 
| [[Image:Chi-Vib6.jpg|thumb|100px| ]] 
| 886 
| 341
| E' 
Stretch - Symmetric and Asymmetric
|-
|}

[[Image:Chi-BCl3_IR.jpg|thumb|600px|left|IR Spectrum of BCl<sub>3</sub> ]]
<br clear="all" />
6 frequencies have been calculated but only 3 peaks can be seen in the spectrum because there are 2 sets of frequencies are degenerate and 1 frequency has zero intensity. The ground state structure has D<sub>3h</sub> symmetry.

=MO Analysis of BH<sub>3</sub>=

BH<sub>3</sub> was  optimised and the MOs were calculated using the B3YLP method the 3-21G basis set. The prodecure followed for [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/2a_optimising_bh3.html optimisation]and [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/5a_molecular_orbitals.html molecular orbital calculation] can be found by clicking on the link.

[http://hdl.handle.net/10042/to-3982 Output Files] for b3lyp/3-21g BH<sub>3</sub> optimisation

[http://hdl.handle.net/10042/to-3961 Output Files] for b3lyp/3-21g BH<sub>3</sub> molecular orbitals

The prodecure followed for [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/5a_molecular_orbitals.html molecular orbital analysis] can be found by clicking on the link.

[[Image:Chi-BH3_MO_Diagram3.jpg|thumb|left|800px|Valence MO Diagram of BH<sub>3</sub> - The 1a'<sub>1</sub> MO is a non-bonding atomic orbital of B 1S<sup>2</sup>.]]
<br clear="all" />

Comparing the MOs in the above MO diagram we can see that there are no significant differences between the real and LCAO MO's. The  quantitative or computed MOs are assumed to be very accurate, at least for the  BH<sub>3</sub> molecule using. This shows that qualitative MO theory is just as accurate  and is very useful as for predicting the MOs of molecules to a relatively good degree of accuracy.

=Transition Metal Complex=

==Optimisation==

The ground state structures of cis and trans Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> were  optimised using the B3YLP method firstly using the LANL2MB basis set, and then again with the LANL2DZ basis set. The prodecure followed for [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/10b_MoC4L2_opt.html optimisation] can be found by clicking on the link.

[http://hdl.handle.net/10042/to-3978 Output Files] for b3lyp/lanl2mb Cis-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> optimisation

[http://hdl.handle.net/10042/to-3979 Output Files] for b3lyp/lanl2mb Trans-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> optimisation

Below are the Jmol of the optimised geometries.

<jmol>
<jmolAppletButton>
<uploadedFileContents>Chi-cis_tm_opt.mol</uploadedFileContents>
<text>Jmol</text>
</jmolAppletButton>
</jmol> of Cis-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub>

<jmol>
<jmolAppletButton>
<uploadedFileContents>Chi-trans_tm_opt.mol</uploadedFileContents>
<text>Jmol</text>
</jmolAppletButton>
</jmol> of Trans-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub>

[http://hdl.handle.net/10042/to-4087 Output Files] for b3lyp/lanl2dz Cis-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> optimisation

[http://hdl.handle.net/10042/to-4088 Output Files] for b3lyp/lanl2dz Trans-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> optimisation

==Frequency Analysis==

[http://hdl.handle.net/10042/to-4093 Output Files] for b3lyp/lanl2mb Cis-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> frequency

[http://hdl.handle.net/10042/to-4094 Output Files] for b3lyp/lanl2mb Trans-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> freqeuncy

The prodecure followed for [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/6_freq_analysis.html vibrational analysis] can be found by clicking on the link.

All of the frequencies were found to be positive showing that the optimisation was sucessfull and a minimum was otained.


{| class="wikitable" border="1"
|+ Carbonyl Stretching Frequencies
! Cis 
|
|
! Trans
|
|
|-
! Frequency/cm<sup>-1</sup>
! Intensity
! Motion
! Frequency/cm<sup>-1</sup>
! Intensity
! Motion
|-
| 1945|| 762|| [[Image:Chi-Cis1.jpg|100px]]Cis|| 1950|| 1475|| [[Image:Chi-Trans1.jpg|100px]] Trans Asymmetric
|-
| 1949|| 1500|| [[Image:Chi-Cis2.jpg|100px]] Trans|| 1951|| 1467|| [[Image:Chi-Trans2.jpg|100px]] Trans Symmetric
|-
| 1959|| 634|| [[Image:Chi-Cis3.jpg|100px]] Asymmetric|| 1977|| 1|| [[Image:Chi-Trans3.jpg|100px]] Asymmetric
|-
| 2024|| 597|| [[Image:Chi-Cis3.jpg|100px]] Symmetric|| 2031|| 4|| [[Image:Chi-Trans3.jpg|100px]] Symmetric
|}

{|align="left"
|[[Image:Chi-Cis_IR.jpg|thumb|left|400px|IR Spectrum of Cis-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> ]]
|[[Image:Chi-Trans_IR.jpg|thumb|left|400px|IR Spectrum of Trans-Mo(CO)<sub>4</sub>(PCl<sub>3</sub>)<sub>2</sub> ]]
|}

Although there are 4 stretching motions there are 2 visible peaks in the spectrum of the cis-isomer because 3 of the stretching motions have similar frequency.
The trans-isomer only has 1 visible peak in the spectrum because 2 of the stretching motions are degenerate and 2 have almost zero intensity.
The computed IR spectra show vibrations at very low energies/frequencies. At room temperature, where the energy is higher, we would expect there to be more vibrations present.
Degeneracy occurs in molecules with high symmetry. The trans isomer has high symmetry and so using this information we can determine which of isomers are cis or trans.

Analysis of the energies of the isomers show that the cis isomer is lower in energy (more stable) than the trans imsomer by a small difference of 1kJ. This shows that the cis isomer should be the preferred isomer. This can be rationalised by stability of the generated dipole when the two PCl<sub>3</sub> groups are cis.. However the literature states that the trans isomer is the most stable. This can be rationalised by the increased stability when there is less steric clash when in the trans conformation. Steric clash is the dominating factor and so hence the trans-isomer is the most stable. The most significant error for the calculation is probabaly the use of Cl atoms instead of phenyl rings. Although Cl atoms have a similar electronic contribution to the bonding they are much less bulky than phenyl rings and hence steric effects would be absent relative to phenyl rings. As with all calculations, using a better basis set would lead to more accurate results.<ref>F. A. Cotton, D. J. Darensbourg, S.. Klein, B. W. S. Kolthammer, Inorg. Chem., 1982, 21 (1), 294–299. [http://dx.doi.org/10.1021/ic00131a055 DOI:10.1021/ic00131a055]</ref><ref>L. Hirsivaara, M. Haukka, J. Pursiainen, Inorg. Chem. Comm. 3 (2000),508–510 [http://dx.doi.org/10.1016/S1387-7003(00)00131-3 DOI:10.1016/S1387-7003(00)00131-3]</ref><ref>D. W. Bennett, T. A. Siddiquee, D.T. Haworth, S.E. Kabir, F. K. Camellia, Journal of Chemical Crystallography, Vol. 34, No. 6, June 2004 ( 2004), 353-359 [http://dx.doi.org/10.1023/B:JOCC.0000028667.12964.28 DOI:http://dx.doi.org/10.1023/B:JOCC.0000028667.12964.28]</ref>

The computed IR spectra show vibrations at very low energies/frequencies. At room temperature, where the energy is higher, we would expect there to be more vibrations present.

=Mini Project=

==Introduction<ref>Andrzej Kochel, Inorganic Chemistry Communications, Volume 10, Issue 12, December 2007, 1440-1443 [http://dx.doi.org/10.1016/j.inoche.2007.09.007 DOI: 10.1016/j.inoche.2007.09.007]</ref>==

In this project I will study the ReCl<sub>3</sub> molecule and its stable and preferred Re<sub>3</sub>Cl<sub>9</sub> cluster.
This project will involve optimisation of the molecules followed by vibrational analysis to confirm optimisation, and molecular orbital/natural bond orbital analysis. The energy of the products will also be looked at to determine whether the molecule or the cluster is more stable.

I will also study TcCl<sub>3</sub> and its Tc<sub>3</sub>Cl<sub>9</sub> cluster and compare this with the Re cluster.

Finally I will study ReBr<sub>3</sub> and its Re<sub>3</sub>Br<sub>9</sub> cluster.

To the right are the optimised structures of ReCl<sub>3</sub> and the Re<sub>3</sub>Cl<sub>9</sub> cluster. The other molecules have the same general structure but with substituted atoms.

{|align="right"
|[[Image:Chi-ReCl3(2).jpg|thumb|150px|Optimised ReCl<sub>3</sub>
<jmol>
<jmolAppletButton>
<uploadedFileContents>ReCl3.mol</uploadedFileContents>
<text>Jmol</text>
</jmolAppletButton>
</jmol>]]
|}
<br clear="all" />
{|align="right"
|[[Image:FilenameRe3Cl9_3DChem.jpg|thumb|200px|Re<sub>3</sub>Cl<sub>9</sub> from 3Dchem.com ]]
|[[Image:Chi-Re3Cl9.jpg|thumb|150px|Optimised Re<sub>3</sub>Cl<sub>9</sub> Cluster
<jmol>
<jmolAppletButton>
<uploadedFileContents>Chi-Re3Cl9.mol</uploadedFileContents>
<text>Jmol</text>
</jmolAppletButton>
</jmol>]]
|}
<br clear="all" />

==Optimisation==

ReCl<sub>3</sub>, Re<sub>3</sub>Cl<sub>9</sub> and the derivatives were optimised using the B3YLP method the  LANL2DZ  basis set. The general prodecure followed for [http://www.ch.ic.ac.uk/hunt/teaching/teaching_comp_lab_year3/2a_optimising_bh3.html optimisation] can be found by clicking on the link. The clusters required 3 optimisations via SCAN before the energy converged. The output files of the final optimised structures are found below:

[http://hdl.handle.net/10042/to-3988 Output Files] for b3lyp/lanl2dz ReCl<sub>3</sub> optimisation

[http://hdl.handle.net/10042/to-3990 Output Files] for b3lyp/lanl2dz Re<sub>3</sub>Cl<sub>9</sub> optimisation

The energy of ReCl<sub>3</sub> is -123.9 Hartrees, and the energy of Re<sub>3</sub>Cl<sub>9</sub> is -372.1 Hatrees. The cluster has a lower energy than 3 individual molecules of ReCl<sub>3</sub> and hence the cluster is preferred and formed.

[http://hdl.handle.net/10042/to-3991 Output Files] for b3lyp/lanl2dz ReBr<sub>3</sub> optimisation

[http://hdl.handle.net/10042/to-3992 Output Files] for b3lyp/lanl2dz Re<sub>3</sub>Br<sub>9</sub> optimisation

In comparison the energy of ReBr<sub>3</sub> is -118.6 Hartrees, and the energy of Re<sub>3</sub>Br<sub>9</sub> is -356.1 Hatrees. In this case the cluster has a higher energy than 3 individual molecules of ReBr<sub>3</sub> and hence the cluster is not formed.

[http://hdl.handle.net/10042/to-3993 Output Files] for b3lyp/lanl2dz TcCl<sub>3</sub> optimisation

[http://hdl.handle.net/10042/to-3994 Output Files] for b3lyp/3-21g Tc<sub>3</sub>Cl<sub>9</sub> optimisation

Similar the energy of TcCl<sub>3</sub> is -124.9 Hartrees, and the energy of Tc<sub>3</sub>Cl<sub>9</sub> is -375.0 Hatrees. Again the cluster has a higher energy than 3 individual molecules of TcCl<sub>3</sub> and hence the cluster is not formed.

==Vibrational Analysis==

Note: The following bond lengths are given in Angstroms, Å, and the stretching frequencies are given in wavenumbers, cm<sup>-1<sup>.

The log file of optimisation only told us that the energy has convereged to zero gradient. Vibrational analysis allows us to determine whether obtimisation to MINIMUM energy has been achieved. Here all the calculated frequencies are positive which shows all optimisations have suceeeded.

[http://hdl.handle.net/10042/to-4122 Output Files] for b3lyp/lanl2dz ReCl<sub>3</sub> frequency

[http://hdl.handle.net/10042/to-4123 Output Files] for b3lyp/lanl2dz Re<sub>3</sub>Cl<sub>9</sub> frequency

{|align=left
|[[Image:Chi-recl3_IR.jpg|thumb|200px|ReCl<sub>3</sub> ]]
|[[Image:Chi-re3cl9_IR.jpg|thumb|200px|Re<sub>3</sub>Cl<sub>9</sub> ]]
|}
<br clear="all" />

===ReCl<sub>3</sub>===
Re-Cl bond length: 2.25

Re-Cl stretch: 349 and 402.

===Re<sub>3</sub>Cl<sub>9</sub>===
Re-Cl bond length: 2.32

Re-Re bond length: 2.42

Re-Cl stretch: 344 (degenerate), 353 (degenerate) and 390

We can see that the Re-Cl bond length has increased in the cluster and hence the stretching frequency has decreased.

<br clear="all" />

[http://hdl.handle.net/10042/to-4124 Output Files] for b3lyp/lanl2dz ReBr<sub>3</sub> frequency

[http://hdl.handle.net/10042/to-4125 Output Files] for b3lyp/lanl2dz Re<sub>3</sub>Br<sub>9</sub> frequency

{|align=left
|[[Image:Chi-rebr3_IR.jpg|thumb|200px| ]]
|[[Image:Chi-re3br9_IR.jpg|thumb|200px| ]]
|}
<br clear="all" />

===ReBr<sub>3</sub>===
Re-Br bond length: 2.43

Re-Br stretch: 241 and 279.

===Re<sub>3</sub>Br<sub>9</sub>===
Re-Br bond length: 2.47

Re-Re bond length: 2.44

Re-Br stretch: 224 (degenerate), 259 (degenerate) and 269

We can see that the Re-Br bond length is greater than the Re-Cl as expected due to Br having more electrons and being larger. Again the Re-Br bond length increases in the cluster however to a lesser extent compared with Re-Cl. The lower stretching frequencies, again, correspond to the increase in bond lengths.

<br clear="all" />

[http://hdl.handle.net/10042/to-4126 Output Files] for b3lyp/lanl2dz TcCl<sub>3</sub> frequency

[http://hdl.handle.net/10042/to-4127 Output Files] for b3lyp/3-21g Tc<sub>3</sub>Cl<sub>9</sub> frequency

{|align=left
|[[Image:Chi-tccl3_IR.jpg|thumb|200px| ]]
|[[Image:Chi-tc3cl9_IR.jpg|thumb|200px| ]]
|}
<br clear="all" />

===TcCl<sub>3</sub>===
Tc-Cl bond length: 2.28

Tc-Cl stretch: 332, 390 and 425.

===Tc<sub>3</sub>Cl<sub>9</sub>===
Tc-Cl bond length: 2.31

Tc-Tc bond length: 2.41

Tc-Cl stretch: 348 (degenerate), 381 (degenerate) and 422

We can see that the Tc-Cl bond length is lower than the Re-Cl as Tc is smaller than Re. It can also be rationalised by molecular orbital overlap; Tc and Cl have similar sized orbitals compared with Re-Cl. Again the Tc-Cl bond length increases in the cluster and to a much lesser extent compared with Re-Cl and Re-Br. Comparison of the stretching frequencies however show no correlation between the bond length and stretching frequencies. I can only suggest that this may be due to inaccuracy of the calculated frequencies or maybe the overall structure of the cluster will have an effect on the stretching frequency eg Tc-Tc bond length.

The prescence of degenerate stretching frequencies tell us that the structure of the cluster must be symmetrical.

<br clear="all" />

==Molecular Orbital Analysis==

[http://hdl.handle.net/10042/to-4128 Output Files] for b3lyp/lanl2dz ReCl<sub>3</sub> MO

[http://hdl.handle.net/10042/to-4129 Output Files] for b3lyp/lanl2dz Re<sub>3</sub>Cl<sub>9</sub> MO

[http://hdl.handle.net/10042/to-4130 Output Files] for b3lyp/lanl2dz ReBr<sub>3</sub> mo

[http://hdl.handle.net/10042/to-4131 Output Files] for b3lyp/lanl2dz Re<sub>3</sub>Br<sub>9</sub> MO

[http://hdl.handle.net/10042/to-4132 Output Files] for b3lyp/lanl2dz TcCl<sub>3</sub> MO

[http://hdl.handle.net/10042/to-4133 Output Files] for b3lyp/3-21g Tc<sub>3</sub>Cl<sub>9</sub> MO

{| class="wikitable" border="1"
|+ ReCl<sub>3</sub> and Re<sub>3</sub>Cl<sub>9</sub> MO
! Molecule
| HOMO
| LUMO
! Cluster
| HOMO
| LUMO
|-
| ReCl<sub>3</sub> || [[Image:Chi-ReCl3_HOMO.jpg|150px]] || [[Image:Chi-ReCl3_LUMO.jpg|150px]] || Re<sub>3</sub>Cl<sub>9</sub> || [[Image:Chi-Re3Cl9_HOMO.jpg|150px]] || [[Image:Chi-Re3Br9_LUMO.jpg|150px]]
|-
| ReBr<sub>3</sub> || Same as ReCl<sub>3</sub> || Same as ReCl<sub>3</sub> || Re<sub>3</sub>Br<sub>9</sub> || [[Image:Chi-Re3Br9_HOMO.jpg|150px]] || [[Image:Chi-Re3Br9_LUMO.jpg|150px]]
|-
| ReBr<sub>3</sub> || Same as ReCl<sub>3</sub> || Same as ReCl<sub>3</sub> || Re<sub>3</sub>Br<sub>9</sub> || Same as Re<sub>3</sub>Cl<sub>9</sub> || Same as Re<sub>3</sub>Cl<sub>9</sub>
|-
|}



{| class="wikitable" border="1"
|+ ReCl<sub>3</sub> and Re<sub>3</sub>Cl<sub>9</sub> NBO
! Molecule
| 
! Cluster
| 
|-
| ReCl<sub>3</sub> || [[Image:Chi-ReCl3_NBO.jpg|200px]] || Re<sub>3</sub>Cl<sub>9</sub> || [[Image:Chi-Re3Cl9_NBO.jpg|200px]]
|-
| ReBr<sub>3</sub> ||  [[Image:Chi-ReBr3_NBO.jpg|200px]] || Re<sub>3</sub>Br<sub>9</sub> || [[Image:Chi-Re3Br9_NBO.jpg|200px]]
|-
| TcCl<sub>3</sub> || [[Image:Chi-TcCl3_NBO.jpg|200px]] || Tc<sub>3</sub>Cl<sub>9</sub> ||  [[Image:Chi-Tc3Cl9_NBO.jpg|200px]]
|-
|}

NBO analysis shows that there is more electron density present on the Re atoms of Re<sub>3</sub>Br<sub>9</sub> relative to those of Re<sub>3</sub>Cl<sub>9</sub>. At the same time the Br atoms are more positively charged than the Cl atoms. This is strongly non prefered in both molecules.

=Conclusion=

The Re<sub>3</sub>Cl<sub>9</sub> cluster is more stable than individual ReCl<sub>3</sub> molecules and hence exist.
Re<sub>3</sub>Br<sub>9</sub> and Tc<sub>3</sub>Br<sub>9</sub> clusters is less stable than individual molecules and hence do no exist.

All optimisations and calculations were sucessful (frequencies were positive) however some molecules required a second or third optimisation before there was convergence. To improve speed and reliability of calculations, the MM2 optimisation method in ChemBio3D was used, and could be used in future, to optimise the geometry of the molecules before they are optimised with Gaussian. In addition it is much easier to draw and edit molecules in Biochem3D than in GaussView.

=References=

<references />