Introduction to Pople-style basis sets

[@lseinjr1]

(For mobile users) 00:52 Electron configurations for typical first, second, and third row elements 03:21 Minimal basis sets [STO-3G] 06:39 Split valence basis sets [3-21G] 12:26 6-31G basis set 16:10 Basis set with polarization functions [6-31G(d)] 22:00 Diffuse functions [6-31+G(d)]

[@fazilat8648]

Thank you sir 😊

[@viragoth_fr]

Thx from France ! I'm searching for a good explication of these kind of basis set and this video give me all I wanted !

[@HeyKevinYT]

Thanks for the explanation. When I heard it in my VASP lecture yesterday, the nomenclature seemed very cryptic. Now it's clear to me.

[@L3tsFlashers]

Hey chemistry-student from Germany. Thank you so much for this video. It helped me so much and now I hope that my experiment on Tuesday is going to be more understandable :).

[@angelamarineau5279]

I have been watching your videos all week. They are so well explained and perfectly line up with everything I am learning. I truly cannot thank you enough for all of the work you have put into these informative videos.

[@lseinjr1]

You are so welcome!

[@user-oi1rw1tl6d]

Thx from Korea. You just saved me swimming in the swamp of Wikipedia... This video is amazing.

[@lseinjr1]

Glad you liked it!

[@DukeofEdinburgh-kp8fp]

another gratitude from Pyongyang.

[@josemiguelrodriguezsantos2744]

Amigo The way you teach is too good and easy to understand. I think your videos will be useful for my computational chemistry course. Thank you so much

[@lseinjr1]

It's my pleasure

[@brokkenwilliam]

Thanks for this! I've been spending the day counting orbitals in input and output files and this has helped me decipher the lingo a lot :)

[@donatocalabrese2960]

This was such a lovely and straightforward explanation! Thank you so much, i can't tell you with words how much I am grateful for your work!

[@lseinjr1]

You're so welcome!

[@ChaoqunMa-d9d]

Thanks for your video! Very good explanation!

[@lseinjr1]

Glad you enjoyed it!

[@PBeigi]

very impressive tutorial. thanks

[@salehalsaee934]

Thanks so much for this informative video with simple and easy way to understand thanks a lot

[@lseinjr1]

Glad it was helpful!

[@subhashishsharma2100]

This was amazing! Thanks a lot

[@drewparmelee3784]

Very good explaination, thank you!

[@lseinjr1]

Glad it was helpful!

[@maryamdaneen7190]

Thank you so much.... basis functions to me are clear now ..yeyyyy👍 I get happy on learning small things and these small things sometimes make a big difference.👍

[@lseinjr1]

You are welcome 😊

[@sankawijewardane4002]

Thank you very much for giving such a valuable explanation.

[@umaturrehman5797]

Amazing video. I just started research in computational chemistry. This video is a great help. Thank you very much Sir.

[@lseinjr1]

Glad it was helpful!

[@ni3cat]

very clear instruction. Thank you for your hard work.

[@lseinjr1]

Glad it was helpful!

[@amalbouzaheur3101]

thank you very much for this straightforward clear lecture

[@lseinjr1]

Glad it was helpful!

[@raziyhebakhshandeh3226]

Thank You for great presentation,really it was helpful.

[@lseinjr1]

You are very kind.

[@Madrinass]

Thank yooouuuuu! Helped me so much

[@lseinjr1]

Glad it helped!

[@athena1234csb]

This video was great! Thank you so much for your help! :)

[@mustakim2144]

Thank you so much for making this video for us!

[@gabrielamaturana5516]

That was a very clear explanation! It was so helpful, thank you very much =)

[@lseinjr1]

Glad it was helpful!

[@armouredlion]

You're the man... Wow, thank you so much

[@lseinjr1]

Glad it helped

[@pfdibvx60g]

شكرا جزيلا لكم

[@МаряКирпа]

it was sooooo useful. thank u so much

[@lseinjr1]

Спасиба!

[@a.aldukhail8696]

Awesome, thanks

[@lseinjr1]

You are very kind.

[@ollinebg]

Thank you very much

[@lseinjr1]

You are welcome!

[@fcsc7583]

Amazingly simple and the most accurate explanation I ever found! Just one doubt, I stumbled upon Karlsruhe basis sets, what would they be? Thank you very much!

[@lseinjr1]

I can give a short answer about Karlsruhe sets. They are designed for use by a certain computational chemistry program (TURBOMOLE). Each basis set has advantages and disadvantages, in that some are more efficient for DFT, for example, and others work better for post-HF calculations. There is an almost infinite variety of ways to use the "diffuse" orbitals ("+" or "++ for Pople-style sets), and the Karlsruhe sets have their own distinctive way of implementing diffuse orbitals.

[@RexGalilae]

How are there two 2s orbitals @8:34? Isn't it just one orbital with 2 electrons?

[@lseinjr1]

No. It was found that if you did the calculations with the orbital scheme that you learned in high school and university, the 1s, 2s, 2p, 3s, 3p, 4s, 3d, and so on, each with exactly one or two electrons, the answers never came out correctly (for molecules). It "should" have worked, but it didn't! To get "correct" molecular orbitals (ones that give energies that match experiments), you have to use "split valences". Let's consider carbon, which has valence electrons in the 2s and 2p orbitals. We know that there is one 2s orbital, and three 2p's - 2pₓ, 2pᵧ, and 2pz. To get good calculations, we have to use a "trick" - we pretend that there are TWO 2s orbitals, each with a slightly different size, and there are two 2pₓ's, each with a different size, and so on. Each carbon atom in the molecule still has six (6) electrons, but those electrons end up in molecular orbitals (no more than two electrons in a molecular orbitals, because of the Pauli Exclusion Principle). Some molecular orbitals use the first 2s, and others use the second 2s and some can use both. This is the idea behind basis sets like 3-21G and 6-31G. It is possible to get even BETTER results by using THREE versions of each valence orbital, as in 6-311G. Eventually, the added benefit of adding more versions is defeated by the extra computer time needed for calculation. While this may seem dubious, there are good theoretical reasons for this technique.

[@yosrasamy2948]

Thanks sooo much

[@PratikKumar_dopamine]

This was helpful. Made it very clear even though I have read about them earlier. So, would you say that (if computationally possible) 6-311++G(d,p) is generally a good place to start with for small organic molecules?

[@lseinjr1]

Yes.

[@DeepakGupta-ce3jv]

Thank you so much sir

[@anya6550]

Thanks for your very clear explanation! I have a question regarding the notation of polarization functions: I have seen notations more complex than (d,p), such as (3df,3pd). I guess that df and pd mean that you are adding two higher angular momentum orbitals (so for a carbon atom, you are adding not only d functions, but also f functions). But what exactly do the numbers mean? I have only seen 2 and 3, and not always together, but also combined, like (3df,2p). I have been searching for info on this matter but I cannot find any explanation. Any help would be appreciated :)

[@lseinjr1]

The notation (3df, 3pd) means that we add three (3) different sized sets of d orbitals, plus one (1) f orbital set to all of the "heavy" atoms (anything heavier than He is "heavy" to a computational chemist). The "3pd" means that we add three (3) differently sized sets of p orbitals and one (1) set of d orbitals to each hydrogen atom. With good theoretical justification, we imagine that if we has an *infinitely* size basis set (set of these orbitals), we would calculate the exact ("true") value. There is always a trade-off between the accuracy of more orbitals, and the expense of the compute time to do the calculation. Does that help?

[@anya6550]

@@lseinjr1 Sure, it does help! I thought that maybe the number had a "hidden" meaning, but now I understand that it indicates the number of different sized set of orbital we're adding. I have not seen a number higher than three, so I am guessing that it is the highest we can go in terms of computation power (right now, at least). Thanks a lot for your answer!

[@HeyKevinYT]

Do people split the core orbitals into different parts like in the valence orbitals?

[@lseinjr1]

No, not usually. They DO tend to model the core orbitals with six (6) Gaussians, whereas in split valence orbitals, each orbital function is modeled by one (1) to three (3) Gaussians.

[@idiotking3505]

Thank you, you wonderful teacher. You might have saved my ass.

[@VijayKumar-zl7fd]

Dear sir I just started my PhD in organic synthesis additionally I will use computation chemistry, so can you suggest me some online course+book for computational which I can use in my PhD.

[@imeneyeddou6596]

You said for diffuse functions it's like a version of S and p orbital super sized ,what about d orbitals?

[@lseinjr1]

Such a thing is certainly possible, but is not usually done. In my personal experience, diffuse orbitals do not seem to make much of a difference. It is generally accepted that diffuse orbitals on hydrogen almost never make a difference. I suspect that diffuse d orbitals would not be worth the trouble to implement. After a certain point, adding more orbitals of any type does not improve the accuracy. To achieve more accurate results, chemists will use methods that are more expensive than density function theory (DFT). These are generally known as post-HF methods, important examples being MP2, MP4, CCSD, and CI.

[@imeneyeddou6596]

@@lseinjr1 thank you so much sir for the answer , your video is very very interesting

[@Lovejassar]

sir which basis set should i use in gaussian software for anionic compounds?

[@lseinjr1]

In general, use the largest basis set possible (based on computer memory, and time available). For the highest accuracy, you want polarization functions, so 6-311G(d) is good, 6-311G(d,p) is better, and then things like 6-311G(3df, 2p) better still. Diffuse functions are necessary for ions, so try 6-311++G(d,p) in place of 6-311G(d,p). In general;, we can use a smaller basis set - like 6-311G(d) for the geometry,- and then use the largest possible set for the energy calculation. A large set makes more difference for the single point energy than for the geometry optimization. One last subtle, but important point. Whichever combination of basis sets you use, one for geometry optimization, and the second for the single point energy, make sure that you use that same combination for *every* molecule and ion in your project. I hope that helps!

[@Lovejassar]

@@lseinjr1 thankyou so much sir for your reply..

[@adupavasista3987]

If we compare STO-3G and 3-21G, overall we have an equal number of GTO i.e., 18 GTO for Iron (valence shell). Then why 3-21G is good compared to the other?

[@lseinjr1]

Great question! "Split valence" basis sets like 3-21G represent each valence orbital by two (2) or more functions. The more functions we use, the more closely we can model the electron distribution in the molecule. 3-21G was not a particularly "good" set, but it proved that the idea of using more than one copy of each valence orbital improved the accuracy. The idea is very similar to the notion of modelling and unknown curve by a polynomial. The greater the order of the polynomial, the wider the selection of curves we could model. The linear mix +b works great if we actually do have a line, but it works poorly for parabolas, for example. Quadratic functions ax2 + bx + c can mode lines and parabolas, but not cubic curves. Splitting each valence orbital into two (2) or (3) three slightly different orbitals has much the same effect. Does this make sense?

[@adupavasista3987]

@@lseinjr1 Great. Thank you.

[@shilpamohanan9288]

sir ,the indication of + sign and use of diffuse function was not clear for Mobile users

[@sadiashaheen3336]

Sir please if you prepare myself for computational ......my exam have started....