Properties of Matter: How Do We Know?

[Avichal]

This maybe a general science question but I'll put it here.

When we say that an element has so-and-so properties, how do we know those are the only properties it has.
For example:- We say an element has properties like electron affinity, electro-negativity etc. How do we know these are the only properties. There could be more, right?

Well, I have an answer and a question to this. Answer is that, we don't know. Only by experimenting and looking at new phenomena we come to know of new properties.
My question is, let's say if we know all the laws of physics perfectly. Since we know an element is made of electrons, protons and neutrons, could we then at least theoretically calculate all the properties?

I don't know why but this question is killing me. I can't sleep!

 

[mfb]

There could be more, right?

And we know there are more. So what? You can even make up your own properties if you like, and measure them. That does not mean everything that can be defined would be useful.

My question is, let's say if we know all the laws of physics perfectly. Since we know an element is made of electrons, protons and neutrons, could we then at least theoretically calculate all the properties?

Theoretically, sure.

Actually, the first assumption is a very good approximation for everyday objects. Particle physics can describe electrons, protons, neutrons and their interactions with incredible precision. Using those laws in systems with several hundred components (large nuclei with many electrons) is still challenging, but that is just a complexity problem, not a fundamental one.

 

[Bystander]

the only properties it has.

You might Google "parachor." An absolute global enumeration of "properties" is more a matter of enumerating user's interests/applications than of enumerating all possibilities for what can be known about an element, or compound. "Parachor" is an example.

 

[DaveC426913]

When we say that an element has so-and-so properties, how do we know those are the only properties it has.
For example:- We say an element has properties like electron affinity, electro-negativity etc. How do we know these are the only properties. There could be more, right?

Note that it depends on where you draw the line.

The set of properties of an individual atom of hydrogen is a very different set than the properties of a mole of hydrogen atoms. Some properties are emergent, and only apply to aggregations.

 

[Avichal]

Some properties are emergent, and only apply to aggregations.

That's what I wanted to ask. Can we theoretically calculate all the properties of an aggregation if we know the properties of its constituents?

As an example:- I was studying about melting point of a substance. Experiments showed that it depends on properties like molecular weight and packing of the molecule. Later by new experiments it was found that melting point also depends on a new property called hydrogen bonding.
So, if we know the laws of physics perfectly, my question is that can we calculate all such properties and not depend on experiments to find out new ones?

You might Google "parachor." An absolute global enumeration of "properties" is more a matter of enumerating user's interests/applications than of enumerating all possibilities for what can be known about an element, or compound. "Parachor" is an example.

Hmm, I think we will need to know what do we exactly mean by "properties", if we want to find all possible properties, right?

 

[Bystander]

need to know what do we exactly mean by "properties"

Yes.

 

[DaveC426913]

That's what I wanted to ask. Can we theoretically calculate all the properties of an aggregation if we know the properties of its constituents?

No.
Imagine taking this list of atoms:
http://web2.airmail.net/uthman/elements_of_body.html
and extrapolating Bach's Toccata and Fugue in D Minor..

 

[mfb]

No.

I think the questions asks about chemical properties, but still:

Why not? A human body and its environment is way beyond the computing capabilities of today's or even tomorrow's computers, but where do you see the fundamental problem that would make such a computation completely impossible if we know the initial state with sufficient precision?

So, if we know the laws of physics perfectly, my question is that can we calculate all such properties and not depend on experiments to find out new ones?

Yes. And we know the laws of physics well enough for all of chemistry. It is "just" a matter of modelling and computing power.

 

[DaveC426913]

I think the questions asks about chemical properties, but still:

Well what is biology - even human biology - but vastly complex chemistry?

Why not? A human body and its environment is way beyond the computing capabilities of today's or even tomorrow's computers, but where do you see the fundamental problem that would make such a computation completely impossible if we know the initial state with sufficient precision?

Sure, in a deterministic Newtonian universe. Not in a quantum universe.

 

[mfb]

Well what is biology - even human biology - but vastly complex chemistry?

Yes, but I think we can agree that "writing music" is not a relevant chemical property of atoms or molecules.

Sure, in a deterministic Newtonian universe. Not in a quantum universe.

The laws of quantum mechanics are deterministic.

 

[DaveC426913]

Yes, but I think we can agree that "writing music" is not a relevant chemical property of atoms or molecules.

We do not agree. The OP posits:

Can we theoretically calculate all the properties of an aggregation if we know the properties of its constituents?

The laws of quantum mechanics are deterministic.

Not sure if we're mincing words, or if I've been reading the wrong books.

Whether or not the laws of QM are deterministic, a quantum universe is not. We cannot predict certain things beyond a certain degree.

 

[mfb]

Whether or not the laws of QM are deterministic, a quantum universe is not.

This depends on the interpretation.
It does not matter for the answer - even if you follow nondeterministic interpretations, a hypothetical supercomputer can calculate all probabilities.

 

[epenguin]

My answer is Science has neither as aim nor method calculating every damn thing.

Nor even explaining every damn thing.

So that relieves it of having to answer "it hasn't explained the flying saucer that several reliable witnesses (airplane pilots etc.) reported seeing in 1948." And similar things creationists will tell you it hasn't explained.

It's not unviersally understood that chemistry for example is not computational physics, though its overlap can be useful, but it's still a science.

When something can be computed, I doubt you can always say it's understood.

About your question I think you should ask yourself - why does it matter.

Without a quantum supercomputer I predict that Evo will soon close this thread.

 

[DaveC426913]

This depends on the interpretation.
It does not matter for the answer - even if you follow nondeterministic interpretations, a hypothetical supercomputer can calculate all probabilities.

But what does 'all probabilities' get you? Is that the same as defining properties?

 

[mfb]

But what does 'all probabilities' get you? Is that the same as defining properties?

Sure. It is the best possible description of the system. What else could you describe?
Actually, it is much better than a measurement, because a measurement will never give you all possible results or their probabilities.

I don't think this side-discussion is very useful.

 

[Avichal]

I found a similar thread on PF:
https://www.physicsforums.com/threads/how-do-physicists-deal-with-emergent-properties.151775/

Great answers which helped me organizing some complicated ideas in my head.

However, I want to ask if the above discussion is present or studied anywhere in scientific literature or are we just presenting our individual theories?

 

[DaveC426913]

However, I want to ask if the above discussion is present or studied anywhere in scientific literature or are we just presenting our individual theories?

Do you mean if there is formal research into - and study of - how we know whether an element has certain properties and only those properties?
I don't think so.

Although, if you're interested in quantum mechanics and spooky action at a distance, Bell's "Hidden Variables" theorem comes close. He refutes the notion that entangled particles at a distance have some sort of hidden property that tells them about each other. This theorem shows that they cannot.