Computational Modelling and Prediction: Mini Computational Project

Overview:

You will carry out a computational study on a molecular system of your choice, using Density Functional Theory at the B3LYP/LANL2DZ level of theory as implemented in Gaussian 16W. In this study, you will use GaussView 6W as the graphical user interface to Gaussian, allowing you to build structures, run calculations and look at results. You should investigate the structure and molecular orbitals of your chosen compound(s) and relate these to spectra and properties. The critical assessment of calculation results is key and you should try to find experimental or calculated data in the literature to compare with your results. You can use your calculations to predict different spectra (IR, NMR, UV/Vis) and might wish to explore other options (e.g. electrostatic potential maps) offered by the software. This should draw on the content covered so far in this unit (Quantum Chemistry), as well as content from the second year Core Chemistry and Practical Chemistry units, along with your knowledge from first year units.

Assessment:

You should present your results in a short, written report, due in week 21 for upload to Blackboard, which will count for 25% of your unit mark. Note that you will also need to complete an online quiz hosted on Blackboard as part of your training on using Gaussian and interpreting outputs, which will count for a further 5% of your unit mark and is due in week 15.

Workplan and Logistics:

See the separate workplan file on Blackboard.

Note there are separate Office Hours/Surgery/Clinic for the Gaussian Individual Project. These are different from the office hours which support the lecture-based material.

It is up to you when you do this work, but support will be provided in weekly drop-in sessions where you can discuss your calculations with postgraduate demonstrators and seek their advice about using GaussView/Gaussian, setting up calculations and interpreting the results. We recommend that you use a PC based in the School of Chemistry for this work if at all possible. The Windows Virtual Desktop can be used to view the results of completed calculations, but it will not allow you to run calculations.

Note that the postgraduate demonstrators are PhD students with considerable expertise in computational chemistry, so we strongly recommend that you engage with them and join office hours for help and support. It would be sensible to do this early on so you can spend more time finalising your report.

Project Design:

You have encountered computational chemistry before. In the first year, you used Entos Envision to look at molecular orbitals and GaussView/Gaussian in experiment 9 (Applications of Computational Chemistry). The second-year practical chemistry unit also includes computational studies using Gaussian, in experiments B3, B8, C7 and C9. Even if you have not done these experiments yourself (or not yet), we suggest that you review them to see how computational studies can be used.

When selecting your own system for study, please bear in mind the following:

- We ask you to use a standard DFT approach, B3LYP/LANL2DZ. The instructions for experiment C9 show you how to set this up

- These DFT calculations are computationally demanding and so we strongly suggest that you limit the size of the molecules you study to 20 atoms or fewer. In consultation with us or the demonstrators, you can stretch this to 30 atoms (we can advise you on pre-optimising structures with lower levels of theory), but please avoid looking at larger systems. In most cases, this will mean that optimisations should not take much longer than around an hour. However, some types of calculations (e.g. TD-DFT) are more demanding and again we suggest that you discuss these with your demonstrators and allow ample time for these to run. You can spend the time while calculations are running on researching the background and analysing other calculation outputs.

- Interpreting calculation results can be easier with molecular symmetry and the initial training will show you some examples of this. When selecting your system, favour those with some symmetry.

- You may not use any of the molecular structures and calculations described on the DLM for this project and the molecules you focus on should have been observed experimentally. As shown in experiment C9, sometimes further calculations on hypothetical structures can help our understanding and you can include these, but please have a known molecule as your central focus.

- We want you to explore GaussView/Gaussian and try out different options (but see advice below!), so please be adventurous, look at the software documentation and discuss your plans and ideas with the demonstrators during the workshop and office hours. As a bare minimum, you should include a geometry optimisation and one other calculation in your report, but we would prefer to see more than that, along with a critical assessment of your results, achieved by comparing them to experimental or calculated data from the literature.

We suggest you try to avoid calculating NMR spectra, periodic calculations and calculation of UV-VIS spectra using TD-DFT as these are computationally more costly and can often be problematic with the Gaussian software.

Report:

Your report should follow a standard structure and contain the following sections:

Abstract – what was done, what are the key outcomes

Introduction – set your project into a broader research context and use this to justify your choice of compound(s) and the data/properties you have selected for calculation.

Computational details – these should give enough information for a trained researcher to repeat your work, meaning software, level of theory, basis set, any additional, non-default settings, all with appropriate references (you can find these in the Gaussian manual). You do not need to give a detailed account of how to build molecules and set up the calculations. A list of the keywords used is not sufficient, though, so please consult appropriate published papers to see how this might be done.

Results and discussion – these should normally be integrated and presented in a single section. You should state what you expect, discuss how your own data compare to expectations, compare with external data (from experiment or calculations) and explain any differences and observations in terms of the relevant theory.

Conclusions – critically assess your study overall and provide suggestions for improvement.

References – use RSC format and include publications from the recent literature context

Appendix – where necessary, this should include any additional data/analysis which is not used directly in the results and discussion section but could be useful to have. Separately, please also submit your *log files with your report.

Please limit your report to 1500-2000 words (shorter reports are fine). Marking will focus on originality, adventure and insight as demonstrated by the chosen system, the quality of your background research and how it is used throughout the report, the choice and interpretation of data for the compounds considered, including your critical assessment, and the clarity of your writing and presentation. We would prefer to see fewer calculations which have been interpreted fully for a wellchosen system, with clear links to material you have studied and researched – try to tell a story - rather than a larger quantity of results which are only poorly understood. Make sure you check your report for clarity, flow and links between sections.

Some possible ideas (amongst others, please use your imagination and we are definitely not being prescriptive here): compare calculated molecular orbital diagrams and bonding including charges with those in standard textbooks; use calculations to assess the accuracy of qualitative arguments on structure and or bonding given in standard textbooks; compare ligands, build up a cluster from say a diatomic and see how properties and bonding change with cluster size; choose a molecule with available spectroscopic data and compare with experiment.

Some Possible Sources for Ideas (amongst others)

Keeler and Wothers “Why Chemical Reaction Happen” and “Chemical Structure and Reactivity”

Albright, Burdett and Whangbo, “Orbital Interactions in Chemistry”

Fleming “Molecular Orbitals and Organic Chemical Reactions“

The Journal of Chemical Education.

For spectroscopic data http:/www.uv-vis-spectral-atlas-mainz.org or a similar database at the

University of Bremen: http://www.iup.physik.uni-bremen.de/gruppen/molspec/index.html