This is my website for my molecular modeling project. The project mainly involves aluminum oxide and its properties which make it a highly useful ceramic. I am also looking into property changes involving the substitution of atoms other than aluminum into aulminum oxide's structure.

Here is a picture of the basic structure of aluminum oxide.

Aluminum Oxide (Al2O3), also known as Alumina, is a highly useful ceramic that is relatively easy to produce and is the strongest and stiffest of the oxide ceramics. Its main properties include: hardness/wear resistance, good thermal conductivity, excellent size and shape capability, and high strength and stiffness. Some common materials that use Alumina are gas laser tubes, wear pads, seal rings, and ballistic armor. Alumina can be found in a number of different formations, but its strongest formation lies within the hexagonal alpha phase. All formations of Aluminum oxide turn to this formation when elevated to high temperatures. The ability to obtain this formation makes it a highly useful and applicable ceramic.

My beginning objective of my project was to compare the strength of aluminum oxide and compare it to a newer material such as carbon fiber. A couple of my classmates, Logan Smestad and Michael Douglas-White, are working on carbon nanotubes so I decided that I would compare test results from Gaussian to see if there were any dramatic differences. The steps I needed to take were quite simple.

First, I would attempt to find a model of Aluminum Oxide and run it through webMO and Gaussian in order to receive some information on energy and bond length.

Second, I would take these results and attempt to interpret them and determine if they were valid.

Next, I would attempt to use the other tools we learned about, such as webAMBER and VMD, in order to receive some interesting models to add to the project.

Then Finally, I would compare results from Logan and Mike's research to explore differences and perhaps determine the stronger material.

As with many experiments, my path to this objective took many different turns due to some road blocks and new findings. So in this next section I will provide some models and pictures that I attempted along the way and explain their contributions.

Here is a picture of the first model I drew and some of the results that it gave. It was a basic Al2O3 molecule.

After conducting this test I decided that I wanted to have a more complex model involving aluminum oxide, since my first test didn't really include the whole structure behind Aluminum Oxide as a ceramic. Here are a few pictures that I attempted to draw myself in WebMO, but these models all failed when I ran tests on them.

When drawing these pictures I started by simply placing the atoms and filling in the bonds, but eventually attempted some 3D images by rotating them and such. This proved to not work so well because all of the pictures I drew failed. :(

Here is an example of a molecule I drew, but webMO ended up changing its geometry into one that didn't seem to make much sense to me. I still took the the data values from this trial and compared them to my first test in order to check validity.

I compared the values between the two trials and they still seem resonable granted that there are a different number of atoms in the molecule.

Another type of model I searched for was one in jmol format. The reason I wanted this was so I could rotate the image of the model and also do many other cool things with it. Alhtough I was never able to import my own jmol image of Aluminum Oxide onto the website, I did find a good jmol image at http://chemagic.com/web_molecules/default.aspx?ID=aluminum_oxide.pdb. I created a link to the website on the picture below.

Within a couple days to the deadline of the project I was able to find a good model of Aluminum Oxide in XYZ format and was able to input it into WebMO. I was also able to somewhat link my project to Jeremy Bock's since the minerals he was studying consisted mostly of Aluminum Oxide. Even though I am dealing with the ceramic side of Aluminum Oxide, I found our link interesting and figured I should add it in with Aluminum Oxide's properties. I also thought that the information I would receive from my last test could be beneficial to his project and mine. As a final note I thought I would mention that Corundum and Aluminum Oxide are completely indentical. Below I have created a link to Jeremy's page so I could show the my newly found characteristics of Aluminum Oxide/Corundum.

Link to Jeremy Bock's Project Page

Here is a picture of the model I pulled of the website in XYZ format and a picture of the webMO configuration

On the night of Thursday January 31, 2007 at around 9:20 p.m. my computation test on the XYZ format picture actually completed. Even when Bert and Ernie had crashed for a few hours due to a tradgic power outage, they managed to work and give me a decent result on Aluminum Oxide. I am glad that Bert and Ernie decided to pull through and hang on after the power outage and help me obtain this calculation.

Here is a picture of the final model obtained and the values of the calculation

After receiving the final model I figured out one of the bond lenghths coming from one of the central Aluminum atoms to one of the Oxygen atoms. The bond was originally 1.720 angstroms and I stretched it to 2.975 angstroms. The beginning energy value for my molecule was -1652.44139244419 kcal/mol. The energy after the stretch was once less than the original value at -950.081054551552 kcal/mol. This test was only run as molecular energy whereas the first was and OPT FREQ therefore I decided to run another test as OPT FREQ and I received an energy value of -5639.52171525. I chose to use my last value as a comparison to carbon nanotubes. Logan's value for his project is located in the link below.

As you can see the difference between Logan's energy values is much, much less than my values. The difference between my energies is about 3987 kcal/mol and the difference between my bond lenghths was 1.255 angstroms. The differences in bond length were relatively close. Actually Logan's molecule was stretched more than mine and it still had a significantly smaller value. This shows that Carbon nanotubes are able to handle more stress than a ceramic such as Aluminum Oxide. This is what I hypothesized near the beginning of the project. So I can now be assured I was correct. I also know that I accomplished something by the end.

The following test was my final result for the project. I was unable to work webAMBER or use VMD. I'm glad I at least got one value that worked out in the end.

Well this was my overall project on Aluminum Oxide and it proved to be quite the journey filled with failures and successes. I would simply like to state that I overall enjoyed this J-term class, even through the frustrations, and that I feel that I really learned some useful skills. These skills being the ability to create a website a write HTML format and also conduct LINUX commands. Learning my way around the chemistry programs here on campus may come in handy in the future too. Thanks a lot for looking at this project and its been fun!

www.accuratus.com/alumox.html

en.wikipedia.org/wiki/Aluminium_oxide

cst-www.nrl.navy.mil/lattice/struk/d5_1.html

chemagic.com/web_molecules/default.aspx?ID=aluminum_oxide.pdb