Where to find experimental data?

[jonjacson]

Hello folks,

When you study a pendulum or the path followed by a ball falling under gravity the first thing you have in mind is the experimental result, because you can actually see it. It is easy to measure or find the position and velocity of those systems. If we talk about a particle with position, mass and velocity we have enough information to predict his movement.

When you study quantum mechanics frequently you read about Hilbert infinite spaces, differential equations, , eigenvalues, eigenvectors, imaginary numbers, integrals, operators, matrix calculus, but, Where is the data?

To make this clear, let's talk about a triatomic molecule, something like our solar system with two planets. Let's say we have 3 protons in the nucleus and 3 electrons. Where can I find the experimental data on this system? What is the total list of observables that is possible to meassure?

I mean, I know from spectroscopy you can measure the difference between energy levels. And from x ray difraction you can learn about its structure, What about the "eigenstates"? Can I measure the "eigenstates"?

We have energy levels, with stern gerlach experiments I guess you can measure spin too. What else is measurable?

My goal is to have first the experimental data, and only then reading about this or that model "hartree fock" and so on.

I don't like to read about a simplifying model before actually knowing the experimental result, it looks completely absurd to me.

 

[DrClaude]

To make this clear, let's talk about a triatomic molecule, something like our solar system with two planets. Let's say we have 3 protons in the nucleus and 3 electrons. Where can I find the experimental data on this system? What is the total list of observables that is possible to measure?

A molecule is not like a solar system, and "3 protons in the nucleus and 3 electrons" don't make a triatomic molecule.

In addition to absorption spectra (mostly IR and VIS-UV), you can have things like mass spectrometry and NMR to gain information about the structure of molecules.

What about the "eigenstates"? Can I measure the "eigenstates"?

No, you can't measure eigenstates directly, only transitions between them. But at one point, the structure of the spectrum gets so complicated that having a model that reproduces it gives you very good confidence that the model is good, especially when it is transferable to a myriad of different molecules.

My goal is to have first the experimental data, and only then reading about this or that model "hartree fock" and so on.

At one point, you have to rely on what others have done before and assume they were correct, otherwise you will not get far. I could show you the ro-vibrational spectrum of Cl₂. What would you do with it? It would simply a wiggly line if you don't already have a model to understand what it can mean.

When I teach the subject, I start by presenting the theory, in a sense asking the students to have confidence in me that I am telling them something correct, then show them some spectra where the features of theory can be observed. But things like the Hartree-Fock method can't easily be "proven" by looking at an experimental spectrum.

I don't like to read about a simplifying model before actually knowing the experimental result, it looks completely absurd to me.

The example you gave was "a pendulum or the path followed by a ball falling under gravity". Do you realize how incredibly much simpler these things are than the spectrum of a molecule? Didn't you at one point accepted that Maxwell's equation were correct, used them to calculate things, and then compare to experiments?

 

[jonjacson]

A molecule is not like a solar system, and "3 protons in the nucleus and 3 electrons" don't make a triatomic molecule.
In addition to absorption spectra (mostly IR and VIS-UV), you can have things like mass spectrometry and NMR to gain information about the structure of molecules.
No, you can't measure eigenstates directly, only transitions between them. But at one point, the structure of the spectrum gets so complicated that having a model that reproduces it gives you very good confidence that the model is good, especially when it is transferable to a myriad of different molecules.
At one point, you have to rely on what others have done before and assume they were correct, otherwise you will not get far. I could show you the ro-vibrational spectrum of Cl₂. What would you do with it? It would simply a wiggly line if you don't already have a model to understand what it can mean.
When I teach the subject, I start by presenting the theory, in a sense asking the students to have confidence in me that I am telling them something correct, then show them some spectra where the features of theory can be observed. But things like the Hartree-Fock method can't easily be "proven" by looking at an experimental spectrum.
The example you gave was "a pendulum or the path followed by a ball falling under gravity". Do you realize how incredibly much simpler these things are than the spectrum of a molecule? Didn't you at one point accepted that Maxwell's equation were correct, used them to calculate things, and then compare to experiments?

I know it is not the same, it is just to put an example on the table.

Thanks for that, (NMR...) I will have a look.

Of course I don't plan to "solve" this problem, it is just that I want to have a look at experimental data.

I realize how simpler it is, but the concepts are always the same. If you have a physical system, experiments, asking nature, or measure it is most important.

Well I readed Faraday before.

 

[DrClaude]

A good textbook will have frequent presentation of experimental results. On example is Demtröder's Atoms, Molecules and Photons.

 

[jonjacson]

It looks interesting. Unfortunately hi has 4 volumes on experimental physics but it is in German. :(

 

[DrClaude]

In the book I mentioned, there are some descriptions of experiments results.

 

[ZapperZ]

Hello folks,

When you study a pendulum or the path followed by a ball falling under gravity the first thing you have in mind is the experimental result, because you can actually see it. It is easy to measure or find the position and velocity of those systems. If we talk about a particle with position, mass and velocity we have enough information to predict his movement.

When you study quantum mechanics frequently you read about Hilbert infinite spaces, differential equations, , eigenvalues, eigenvectors, imaginary numbers, integrals, operators, matrix calculus, but, Where is the data?

To make this clear, let's talk about a triatomic molecule, something like our solar system with two planets. Let's say we have 3 protons in the nucleus and 3 electrons. Where can I find the experimental data on this system? What is the total list of observables that is possible to meassure?

I mean, I know from spectroscopy you can measure the difference between energy levels. And from x ray difraction you can learn about its structure, What about the "eigenstates"? Can I measure the "eigenstates"?

We have energy levels, with stern gerlach experiments I guess you can measure spin too. What else is measurable?

My goal is to have first the experimental data, and only then reading about this or that model "hartree fock" and so on.

I don't like to read about a simplifying model before actually knowing the experimental result, it looks completely absurd to me.

There appears to be a fallacy that one can actually make sense of "experimental data" without actually knowing (i) the actual physics surrounding the experimental technique being used and (ii) the physics that is relevant to that data. This is no different than amateurs clamoring to want to see data on climate change, thinking that a conclusion can be THAT obvious simply by looking at the data without understanding the science behind it.

For example, if I give you the spectroscopy done for NH₃ (ammonia), will you actually be able to pick out the effect that is a direct consequence of the superposition of states for the N atom in the molecule?

As for the answer to your question, why can't you google for it? Have you even done a search on various journals? Have you done Google Scholar search?

Zz.

 

[jonjacson]

There appears to be a fallacy that one can actually make sense of "experimental data" without actually knowing (i) the actual physics surrounding the experimental technique being used and (ii) the physics that is relevant to that data. This is no different than amateurs clamoring to want to see data on climate change, thinking that a conclusion can be THAT obvious simply by looking at the data without understanding the science behind it.

For example, if I give you the spectroscopy done for NH₃ (ammonia), will you actually be able to pick out the effect that is a direct consequence of the superposition of states for the N atom in the molecule?

As for the answer to your question, why can't you google for it? Have you even done a search on various journals? Have you done Google Scholar search?

Zz.

Why is that I only get negative answers and critics when I ask for simple experimental data? Is this normal?

XXi century physics, the physics without real data.

Of course I have done a research but I am not sure if I have found the whole complete set of data that is possible to be observed.

 

[ZapperZ]

Why is that I only get negative answers and critics when I ask for simple experimental data? Is this normal?

XXi century physics, the physics without real data.

Of course I have done a research but I am not sure if I have found the whole complete set of data that is possible to be observed.

First of all, let me be VERY clear to you on who I am. I am an experimentalist! So before you go off on the deep end about your ignorant idea about physics, pay attention to what I was telling you!

Secondly, there is no such thing as a "complete set of data"! You look for the property that you want and go find the relevant experiments that measure THAT property. And note that I said "experiments". Very often, several different experiments measure the SAME thing in different ways.

And no, I have no idea what kind of "research" you have done and on what, because you didn't give a list of what you found.

BTW, if I give you this:

https://arxiv.org/pdf/cond-mat/9904449.pdf

and tell you that Fig. 1 and 2 are the "data", how would you find this of any use? Are you under the impression that people will do a dump of their tons and tons of raw numbers online or in their publication? Is this what you mean by "data"?

Zz.