Periodic Table/Quantum M.
Eric Scerri
Michael S. Morris wrote:
> Thursday, the 15th of October, 1998
>
> Eric Scerri:
> I hope other members of this group
> will consider reading the article in
> Scientific American (Sept.98) for themselves
> even though it has received such a bad
> press from Lewis Mammel.
>
> I will certainly consider reading it, along with
> many other things I already consider reading. But,
> I'm afraid my sentiments here are with Lew. He
> says he's quoting you, and he's normally a pretty
> darn accurate sort:
> Despite the efforts of many
> physicists and chemists, quantum
> mechanics cannot explain the
> periodic table any further. For
> example, it cannot explain from
> first principles the order in
> which electrons fill the various
> shells.
>
> You did write this, did you not?
> The problem lies in the distinction
> Lew is quite correct in pointing to
> between this lack of explanation being
> a problem with quantum mechanics versus
> being a shortcoming of calculation. It *is* a
> shortcoming of calculation. It is *not*
> a problem with quantum mechanics (i.e. no one
> in his right mind is ready to jettison
> quantum mechanics because the filling
> of the orbitals in copper has not yet
> been calculated from first pronciples).
> The way you worded it implies it is a
> problem with quantum mechanics and not
> a question of calculation.
>
> Now, I'm all willing to believe you know your
> stuff, and to take your rec that the rest of your article
> is impeccable. I'm also willing to infer that
> you were writing more about The Periodic Table
> than about the explanatory power of quantum
> per se. In other words, that you might just as
> well have written "calculation" instead of "quantum
> mechanics". Also, I quite agree with your posing
> such calculation as a challenge---there may even
> be something to be learned from it. But, I'm
> a theoretical physicist and beg to differ if it
> is your contention that quantum mechanics *doesn't*
> explain the orbital fillings of copper. There is
> no reason whatsoever for me to doubt that it does.
>
> All such belief (as I express here) in science of
> course is corrigible---it *could* be that there would
> be something big to be learned about quantum
> mechanics by trying to calculate copper from first
> principles. But I sure as heck doubt it.
>
> Mike Morris
> (msmo...@netdirect.net)
----------------------------------------------------
Thank you for your response Mike. Finally somebody who seems to know
what they are talking about and does not think I am a pomo or a ctritic
of science etc.etc.
You are correct in saying that I did not express myself as well as I
might have in what I said regarding QM and the periodic table.
However, I am sure you appreciate that in writing for Scientific
American one takes the risk of having every single word distorted in the
interest of greater clarity for the average reader.
The editors re-wrote large sections of my piece several times. It was a
constant struggle to try to keep some of the original version intact.
The quantum mechanical section came off worse of all. As I think I
mentioned I have a better version in American Scientist, Nov-Dec, 97.
Now to your specific points. I repeat I am not saying that quantum
mechanics cannot predict say the configuration of copper from first
principles. All I am saying, never mind the specific wording now, is
that nobody has yet used QM to perform such a calculation successfully.
Actually Melrose and I come close in another paper in Journal of
Chemical Education.
Now it seems to me that you are somewhat overimpressed by the ability of
quantum mechanics to explain such things as the configuration of copper
even though nobody has done it. maybe it's time physicists attempted it
instead of concentrating on the more esoteric work that we always hear
about.
I agree it could be done but then again if quantum mechanics is going to
break down someday, as I am sure you agree it will eventually, then why
not in something as humble as copper?
Who would have thought that when Planck was poking around in black body
radiation it would lead to a revolution in physics?
But my ambitions are more local. If physics claims to explain the
periodic table it should have by now have predicted exactly where the
elements recur without recourse to semi-empirical procedures. Or am I
asking for too much?
Eric Scerri
Eric Scerri
Paul J. Gans wrote:
> David Iain Greig (gr...@ediacara.org) wrote:
> >Absonite <abso...@aol.comNoBull> wrote:
> >>>Subject: Re: My article in Scientific American (Periodic Table)
> >>>From: harshman....@sjm.infi.net (John Harshman)
> >>>
> >>>Why the concentration on 7, anyway? As far as I can tell, the
period table
> >>>repeats mostly in 8's. The obvious reason for that is that valence
> >>>electrons occupy the s and p orbitals of successive electron
shells, and
> >>>the s and p orbitals combined hold a maximum of 8 electrons.
Simple, isn't
> >>>it?
> >>
> >>It seems quite apparent harshman that you either know nothing about
the 100
> >>elements or you simply have no idea of what you are talking about.
>
> >Since I am holding a Periodic Table with 107 elements on it, you seem
to
> >be out of date with yours.
>
> >101 - Mendelevium
> >102 - Nobelium
> >103 - Lawrencium
>
> >and the pedestrianly named 104-107.
>
> >And the periodic table does repeat in 8's.
>
> >--D.
> >>
>
> Hmm. First it repeats in twos. Then it repeats in eights.
> Then it repeats in 18's, and then the repeat becomes 32. And
> if we had even more elements the next repeat would be 48.
>
> I have *no* idea where this eight business comes from,
> except possibly 1904 textbooks. ;-)
>
> Now, for the folks who want to know where chemical information
> comes from (you know the breed: "evolution is impossible
> because it requires information to be created from nothing.")
> it is all coded into Schroedinger's Equation with one or
> two additions supplied by Pauli.
>
> ----- Paul J. Gans [ga...@panix.com]
OK. At least this is a little better but here are a few comments.
1. The next repeat will not be after 48 but 50. This is elementary.
There
are 9 g orbitals making a possible total of 18 further electrons. 32 +
18 is 50
not 48!
2.It all comes "in principle" from Schrodinger's equation and Pauli's
Principle
but until this has been demonstrated rigorously it is only an in
principle
hope. And before we all start on a completely different tangent I am
not trying
to let creationism in by the back door or anything. I am not a pomo.
Nor am I
citiquing science in general.
8. Where does the 8 come from. Some of the older periodic tables for
example
Newlands who predated Mendeleev spoke of a law of octaves. In some
senses the
order periodic tables convery information that the satndard medium-long
form
does not. For exampl mendeleev groups together the alkali metals with
the noble
metals (Cu, Ag, Au) since they all show valencies of 1. But it's more
subtle
than that becuase he offsets the noble metals in a column within group
one.
Read my article. Sci Am Sept, 98, 56-61.
Are you the same Gans who has published articles with Hans Primas?
Eric Scerri
wf...@enter.netxx
wrote:
>
>Now to your specific points. I repeat I am not saying that quantum
>mechanics cannot predict say the configuration of copper from first
>principles. All I am saying, never mind the specific wording now, is
>that nobody has yet used QM to perform such a calculation successfully.
>Actually Melrose and I come close in another paper in Journal of
>Chemical Education.
>
>Now it seems to me that you are somewhat overimpressed by the ability of
>
>quantum mechanics to explain such things as the configuration of copper
>even though nobody has done it. maybe it's time physicists attempted it
>
interesting comment. your comments on this, from the Nobel prize for
chemistry (won by 2 professors from a former alma mater,
carnegie-mellon)? (quote is from the Nobel committee's press release,
available thru CNN):
>The laws of quantum mechanics as formulated more than 70 years ago make it theoretically
> possible to understand and calculate how electrons and atomic nuclei interact to build up
> matter in all its forms. The task of quantum chemistry is to exploit this knowledge to
> describe the molecular system. This has proved easier said than done. It was not until the
> beginning of the 1960s that development really started, when two events became decisive.
> One was the development of an entirely new theory for describing the spatial distribution
> of electrons, and the other was the use of the increasing potential offered by the computer.
> Walter Kohn showed in 1964 that the total energy for a system described by the laws of
> quantum mechanics can be theoretically calculated if the electrons' spatial distribution
> (electron density) is known. The question is only how the energy depends on the density.
> Kohn gave important clues based on what this dependence looked like in an imaginary
> system with free electrons. It was to take several decades and contributions from many
> researchers, however, before the equation for determining the energy was sufficiently
> accurately mapped to permit large-scale studies of molecular systems. This has taken place
> partly through the adaptation of a small number of variables to experimental data. The
> method Kohn introduced came to be known as the density-functional theory. It is now used
> in studies of numerous chemical problem areas, from calculating the geometrical structure
> of molecules (i.e. bonding distance and angles) to mapping chemical reactions.
tpham...@my-dejanews.com
scra...@earthlink.net wrote:
> Michael S. Morris wrote:
>
> > Thursday, the 15th of October, 1998
> > Eric Scerri:
> > Despite the efforts of many
> > physicists and chemists, quantum
> > mechanics cannot explain the
> > periodic table any further. For
> > example, it cannot explain from
> > first principles the order in
> > which electrons fill the various
> > shells.
> > You did write this, did you not?
> > The problem lies in the distinction
> > Lew is quite correct in pointing to
> > between this lack of explanation being
> > a problem with quantum mechanics versus
> > being a shortcoming of calculation. It *is* a
> > shortcoming of calculation. It is *not*
> > a problem with quantum mechanics (i.e. no one
> > in his right mind is ready to jettison
> > quantum mechanics because the filling
> > of the orbitals in copper has not yet
> > been calculated from first pronciples).
> > The way you worded it implies it is a
> > problem with quantum mechanics and not
> > a question of calculation.
The problem is with the meaning of the word "explain".
I always regard it as a qualitative understanding.
That is, the origins of orbitals is explained,
but computing the lowest energy configuration may not be
easy due to subtle effects.
[snip]
> Thank you for your response Mike. Finally somebody who seems to know
> what they are talking about and does not think I am a pomo or a ctritic
> of science etc.etc.
Pomo? [shudder]
[snip]
> Now to your specific points. I repeat I am not saying that quantum
> mechanics cannot predict say the configuration of copper from first
> principles. All I am saying, never mind the specific wording now, is
> that nobody has yet used QM to perform such a calculation successfully.
> Actually Melrose and I come close in another paper in Journal of
> Chemical Education.
[snip]
I was unaware that it was that difficult. I may give it a whirl
with an acquaintance or two. First I have to read the J. Chem Ed.
paper.
-----------------------------------------------------------------------
Tracy P. Hamilton |O Beer! O Hodgson, Guinness, Alsopp, Bass!
Building Manager, Alco Hall |Names that should be on every infants' tongue!
University of Ediacara | - Charles Stuart Calverly
-----------------------------------------------------------------------
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Don Lowe
<scra...@earthlink.net> writes:
>
>Michael S. Morris wrote:
>
>> Thursday, the 15th of October, 1998
>>
>> Eric Scerri:
>> I hope other members of this group
>> will consider reading the article in
>> Scientific American (Sept.98) for themselves
>> even though it has received such a bad
>> press from Lewis Mammel.
>>
>> I will certainly consider reading it, along with
>> many other things I already consider reading. But,
>> I'm afraid my sentiments here are with Lew. He
>> says he's quoting you, and he's normally a pretty
>> darn accurate sort:
>> Despite the efforts of many
>> physicists and chemists, quantum
>> mechanics cannot explain the
>> periodic table any further. For
>> example, it cannot explain from
>> first principles the order in
>> which electrons fill the various
>> shells.
>>
>> You did write this, did you not?
>> The problem lies in the distinction
>> Lew is quite correct in pointing to
>> between this lack of explanation being
>> a problem with quantum mechanics versus
>> being a shortcoming of calculation. It *is* a
>> shortcoming of calculation. It is *not*
>> a problem with quantum mechanics (i.e. no one
>> in his right mind is ready to jettison
>> quantum mechanics because the filling
>> of the orbitals in copper has not yet
>> been calculated from first pronciples).
>> The way you worded it implies it is a
>> problem with quantum mechanics and not
>> a question of calculation.
>>
>> Now, I'm all willing to believe you know your
>> stuff, and to take your rec that the rest of your article
>> is impeccable. I'm also willing to infer that
>> you were writing more about The Periodic Table
>> than about the explanatory power of quantum
>> per se. In other words, that you might just as
>> well have written "calculation" instead of "quantum
>> mechanics". Also, I quite agree with your posing
>> such calculation as a challenge---there may even
>> be something to be learned from it. But, I'm
>> a theoretical physicist and beg to differ if it
>> is your contention that quantum mechanics *doesn't*
>> explain the orbital fillings of copper. There is
>> no reason whatsoever for me to doubt that it does.
>>
>> All such belief (as I express here) in science of
>> course is corrigible---it *could* be that there would
>> be something big to be learned about quantum
>> mechanics by trying to calculate copper from first
>> principles. But I sure as heck doubt it.
>>
>> Mike Morris
>> (msmo...@netdirect.net)
>
>----------------------------------------------------
>
<snip>
>Now to your specific points. I repeat I am not saying that quantum
>mechanics cannot predict say the configuration of copper from first
>principles. All I am saying, never mind the specific wording now, is
>that nobody has yet used QM to perform such a calculation successfully.
>Actually Melrose and I come close in another paper in Journal of
>Chemical Education.
So you are saying its a difficult problem.
>Now it seems to me that you are somewhat overimpressed by the ability of
>quantum mechanics to explain such things as the configuration of copper
>even though nobody has done it.
There are many difficult problems in QM. That doesn't stop me from
being impressed by its all encompassing ability to explain many
physical phenomena.
>maybe it's time physicists attempted it
>instead of concentrating on the more esoteric work that we always hear
>about.
This seems to be an extremely cheap shot.
What is esoteric about ceramic toughening, fuel cells, atomic absorption
spectroscopy, ore extraction, computer tomography, superconductivity,
metal vapour lasers and acoustics, all of which is studied in our
"physics department"?
Perhaps if you have some funding for such a job (since funding tends
to dictate what is explored these days) you should employ an unemployed
theoretical physicist to work on this problem instead of complaining
that no-one has done it.
>I agree it could be done but then again if quantum mechanics is going to
>break down someday, as I am sure you agree it will eventually, then why
>not in something as humble as copper?
Perhaps but I doubt it ... and maybe that's why people aren't that
interested in it ... its probably not considered a fruitful place
to look for QM anomalies. However, as I said, if you have the
funding, I am sure you can interest someone.
>Who would have thought that when Planck was poking around in black body
>radiation it would lead to a revolution in physics?
Well Plank was explaining a clear discrepancy between experiment and
the best available models. Even though an exact solution does not appear to be
availble I don't think it could be called a *clea discreoency) since the
principles of QM
>
>But my ambitions are more local. If physics claims to explain the
>periodic table it should have by now have predicted exactly where the
>elements recur without recourse to semi-empirical procedures. Or am I
>asking for too much?
Your asking alot if you expect people to do it for free. I am guessing
(so I could be wrong) that such a problem will require many hours
on a supercomputer. It cannot be done analytically so must be performed
numerically. Its a many-body problem in which all the interactions
must be accounted for.
One last comment ... why is this in talk.origins and not the
appropriate sci.* group?
>
>Eric Scerri
Don
(remove the obvious for email)
Don Lowe
Hoepfully it wasn't a substantive problem since I was only responding
to a small point at the time ... however, I shall remedy the offending
bit below.
In article <70c4rm$hbo$1...@towncrier.cc.monash.edu.au>, Don Lowe
<don....@DUMMYsci.monash.edu.au> writes:
>In article <3627A1FB...@earthlink.net>, Eric Scerri
><scra...@earthlink.net> writes:
<snip>
>>Who would have thought that when Planck was poking around in black body
>>radiation it would lead to a revolution in physics?
And I was writing when I accidently pressed send :-)
>Well Plank was explaining a clear discrepancy between experiment
>and the best available models. Even though an exact solution does
>not appear to be availble I don't think it could be called a *clea
>discreoency) since the principles of QM
This should read:
"Well, Plank was explaining a clear discrepancy between experiment
and the best available models (classical mechanics). Even though
an exact solution does not appear to be currently available I don't
think it could be called a *clear* discrepency since the principles
of QM are very successful at the atomic level."
<snip>
>>Eric Scerri
>
>Don
>(remove the obvious for email)
Don (again)
Paul J. Gans
>Michael S. Morris wrote:
>----------------------------------------------------
>Thank you for your response Mike. Finally somebody who seems to know
>what they are talking about and does not think I am a pomo or a ctritic
>of science etc.etc.
>You are correct in saying that I did not express myself as well as I
>might have in what I said regarding QM and the periodic table.
>However, I am sure you appreciate that in writing for Scientific
>American one takes the risk of having every single word distorted in the
>interest of greater clarity for the average reader.
>The editors re-wrote large sections of my piece several times. It was a
>constant struggle to try to keep some of the original version intact.
>The quantum mechanical section came off worse of all. As I think I
>mentioned I have a better version in American Scientist, Nov-Dec, 97.
>Now to your specific points. I repeat I am not saying that quantum
>mechanics cannot predict say the configuration of copper from first
>principles. All I am saying, never mind the specific wording now, is
>that nobody has yet used QM to perform such a calculation successfully.
>Actually Melrose and I come close in another paper in Journal of
>Chemical Education.
>Now it seems to me that you are somewhat overimpressed by the ability of
>quantum mechanics to explain such things as the configuration of copper
>even though nobody has done it. maybe it's time physicists attempted it
>instead of concentrating on the more esoteric work that we always hear
>about.
>I agree it could be done but then again if quantum mechanics is going to
>break down someday, as I am sure you agree it will eventually, then why
>not in something as humble as copper?
>Who would have thought that when Planck was poking around in black body
>radiation it would lead to a revolution in physics?
>But my ambitions are more local. If physics claims to explain the
>periodic table it should have by now have predicted exactly where the
>elements recur without recourse to semi-empirical procedures. Or am I
>asking for too much?
>Eric Scerri
Let me add to this overlong post. Again, I'm not sure
that non-chemists and non-physicists will quite follow.
There is no known or suspected problem with quantum
mechanics per se in terms of computing the electronic
situation in copper (or uranium, for that matter).
There are a number of serious computational difficulties
that keep such calculations from being done. Yes, we
know how to do them, and yes, in every case where they
have been done, we get the expected results.
The problem is not just that it takes computer time.
The problem is that atoms with more than one electron
(which, IIRC is most of them ;-) do not really have neat
"orbitals" or "energy levels" into which one can
compartmentalize them. Yes, chemists do use language
stolen from the structure of the hydrogen atom)
to describe the electronic structure of such atoms,
but that doesn't mean that there really is a 3s
electron in copper that is distinguishable, ab
initio, from a 3p electron in the same atom. Put
another way, there are no sets of "pure" electronic
states for multi-electron atoms.
Further, copper is in a position of known difficulty
in the periodic chart. It has (or "should" have)
9 electrons in the 3-d subshell. Or so we would
have you believe when you take general chemistry.
Then we tell you that since copper is only one electron
short of having a "complete" d subshell, it "borrows"
an electron from the 4s subshell and completes the
d subshell. And when we tell you this we wave our
arms around and talk about spectroscopic evidence and
Hund's rules. Students (quite properly) get glassy-eyed
at this.
Can we explain this via quantum mechanics? As Eric
Scerri has pointed out, no, we cannot.
It is something like asking "how does the stork
manage to carry a 12 pound baby through the air?"
The answer is that a stork does no such thing. And
the answer for copper is that there really isn't any
such thing as 4s or 3d electrons. They are a
convenient fiction. More technically, there aren't
any "good" quantum numbers for such things in
multi-electron atoms.
What one can measure, for instance, is the ionization
potential for copper. That is, the minumum energy needed to
remove an electron from a copper atom. *That* can be
computed via quantum mechanics (I don't know if it has
been done _ab initio_) but everyone expects that it would,
if done, give the right answer.
Since I've gone on at such length, let me add one more
complication. In really heavy atoms one must worry about
relativistic effects. This exponentially multiplies the
computational difficulties to the point where such
computations are very very difficult.
Executive Summary: You won't solve the quantum mechanical
equations for copper to get the electron-configuration
results you want because that's the wrong question to ask.
Paul J. Gans
>Paul J. Gans wrote:
[...]]
>> Hmm. First it repeats in twos. Then it repeats in eights.
>> Then it repeats in 18's, and then the repeat becomes 32. And
>> if we had even more elements the next repeat would be 48.
>>
>> I have *no* idea where this eight business comes from,
>> except possibly 1904 textbooks. ;-)
>>
>> Now, for the folks who want to know where chemical information
>> comes from (you know the breed: "evolution is impossible
>> because it requires information to be created from nothing.")
>> it is all coded into Schroedinger's Equation with one or
>> two additions supplied by Pauli.
>>
>> ----- Paul J. Gans [ga...@panix.com]
>OK. At least this is a little better but here are a few comments.
>1. The next repeat will not be after 48 but 50. This is elementary.
Yes, I can't add in my head.
>There
>are 9 g orbitals making a possible total of 18 further electrons. 32 +
>18 is 50
>not 48!
Right. Mea culpa.
>2.It all comes "in principle" from Schrodinger's equation and Pauli's
>Principle
>but until this has been demonstrated rigorously it is only an in
>principle
>hope. And before we all start on a completely different tangent I am
>not trying
>to let creationism in by the back door or anything. I am not a pomo.
>Nor am I
>citiquing science in general.
I don't totally agree. First, it is not possible to test
every single possible result. If quantum mechanics failed
to produce the correct *computable* results, the foundations
of modern physics would shake.
By *computable* result I mean the answer to a quantum-mechanically
meaningful question. The distance of an electron from the nucleus
and the size of an electron are, for instance, NOT meaningful
questions in quatum mechanics.
>8. Where does the 8 come from. Some of the older periodic tables for
>example
>Newlands who predated Mendeleev spoke of a law of octaves. In some
>senses the
>order periodic tables convery information that the satndard medium-long
>form
>does not. For exampl mendeleev groups together the alkali metals with
>the noble
>metals (Cu, Ag, Au) since they all show valencies of 1. But it's more
>subtle
>than that becuase he offsets the noble metals in a column within group
>one.
>Read my article. Sci Am Sept, 98, 56-61.
I have and a good one it is. But the idea of octaves is long
out of date and no particular good comes out of using it unless,
as you were doing, there is a historical point to be made.
>Are you the same Gans who has published articles with Hans Primas?
No, I'm not. There is a fairly well-known Paul Gans in England
with whom I'm sometimes confused, to my benefit, I might add.. ;-)
Robert Grumbine
A nice post which clarified just why there was a serious question
involved (a point which was bothering me, as I'd recently read Herzberg's
Atomic Spectra and Atomic Structure, and he seemed to think things were
pretty well managed).
My non-serious addition:
>The problem is that atoms with more than one electron
>(which, IIRC is most of them ;-)
Not by number! Most (ca. 92% by number, 75% by weight) of
the atoms in the universe only have 1 electron.
:-)
--
Robert Grumbine rm...@access.digex.net http://www.access.digex.net/~rmg3/
Sagredo (Galileo Galilei) "You present these recondite matters with too much
evidence and ease; this great facility makes them less appreciated than they
would be had they been presented in a more abstruse manner." Two New Sciences
Paul J. Gans
>In article <70g34c$4...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
> A nice post which clarified just why there was a serious question
>involved (a point which was bothering me, as I'd recently read Herzberg's
>Atomic Spectra and Atomic Structure, and he seemed to think things were
>pretty well managed).
> My non-serious addition:
>>The problem is that atoms with more than one electron
>>(which, IIRC is most of them ;-)
> Not by number! Most (ca. 92% by number, 75% by weight) of
>the atoms in the universe only have 1 electron.
>:-)
>--
>Robert Grumbine rm...@access.digex.net http://www.access.digex.net/~rmg3/
>Sagredo (Galileo Galilei) "You present these recondite matters with too much
>evidence and ease; this great facility makes them less appreciated than they
>would be had they been presented in a more abstruse manner." Two New Sciences
Well, there you go. So we can analytically solve Schroedinger's
equation for 92% of the atoms in the universe. Since we can
also do a *fine* _ab initio_ job on helium, which probably makes
up a significant amount of the remaining 8%, other atoms are
just "noise" and can be safely ignored. Quantum mechanics
accurately explains 95+% of the atoms in the universe.
------ Paul J. Gans [ga...@panix.com]
PS: Do I *need* a smiley?
Robert Parson
>Well, there you go. So we can analytically solve Schroedinger's
>equation for 92% of the atoms in the universe. Since we can
>also do a *fine* _ab initio_ job on helium, which probably makes
>up a significant amount of the remaining 8%, other atoms are
>just "noise" and can be safely ignored. Quantum mechanics
This reminds of a beautiful, but little-known, triumph of
quantum mechanics: the H^- negative ion, two electrons and one proton.
It has exactly one bound state, but nevertheless plays a major role
in the solar spectrum. I believe Bethe first predicted its stability
in the late 1920's or early 1930's. In the 1940's Chandrasekhar
achieved a very accurate result for the binding energy (bet you never
knew Chandrasekhar did ab initio quantum chemistry - that guy could
do _anything_. Of course he did the laborious numerical calculations
entirely by hand.) H^- may be the first example of a chemical species
whose existence was predicted by quantum mechanics before it was
observed in the laboratory (as an isolated gas-phase ion, at any rate.)
------
Robert
MJR
>
> In article <3627A1FB...@earthlink.net>, Eric Scerri
> <scra...@earthlink.net> writes:
[ enormous snip ]
> >Now it seems to me that you are somewhat overimpressed by the ability of
> >quantum mechanics to explain such things as the configuration of copper
> >even though nobody has done it.
What the hell are you talking about? The electronic structure of
atoms was solved seventy years ago.
>
> There are many difficult problems in QM. That doesn't stop me from
> being impressed by its all encompassing ability to explain many
> physical phenomena.
One of the interesting features of QM is that anyone, with a little
real exposure to the field, can pose problems that are impossibly
difficult to solve. The trick lies in posing problems that i) we can
get some kind of solution, if we are clever and ii) have some
interesting feature to the solution.
>
> >maybe it's time physicists attempted it
> >instead of concentrating on the more esoteric work that we always hear
> >about.
>
> This seems to be an extremely cheap shot.
>
> What is esoteric about ceramic toughening, fuel cells, atomic absorption
> spectroscopy, ore extraction, computer tomography, superconductivity,
> metal vapour lasers and acoustics, all of which is studied in our
> "physics department"?
>
> Perhaps if you have some funding for such a job (since funding tends
> to dictate what is explored these days) you should employ an unemployed
> theoretical physicist to work on this problem instead of complaining
> that no-one has done it.
>
It seems to me that Mr Scerri doesn't understand the fundamental
dynamic of physics research: we attack problems because there seems to
be some hope that they can be conquered, not neccessarily because there
is some direct and obvious "importance" to the solution. This is a
practical reality in the field, which is justified by the fact that
important and useful results spring from organising research in this
way. Remembering back to Senator Proxmire and his Golden Fleece award,
if we refused funding to all projects except those that promised a 450
MPG carburetor *tomorrow* we would soon reduce the scientific enterprise
to rubble.
Paul J. Gans
> ------
> Robert
No, I didn't know that. It's a great story!
Robert Parson
>Robert Parson (rpa...@spotNO.SPAMColorado.edu) wrote:
>
>> This reminds of a beautiful, but little-known, triumph of
>> quantum mechanics: the H^- negative ion, two electrons and one proton.
>
>No, I didn't know that. It's a great story!
Unfortunately I garbled it somewhat (posting from memory.) Here's
the correct story: Bethe first predicted the stability of H^- in
1929, using a three-parameter trial wave function containing an
explicit interelectron repulsion term. Hylleraas in 1930 and Henrich
in 1943 improved the calculation by using 6 and 11-parameter functions,
respectively, getting an ionization potential of about 0.75 eV.
(Correlation is absolutely essential in this problem; Hartree-Fock
predicts that the ion is unbound, and you have to do a whopping big
CI expansion to get an answer as good as Henrich's. Pictorially, the
ion is stable because the two electrons spend a lot of time on opposite
sides of the nucleus.) Chandrasekhar then worked out the intensities
of bound-bound and bound-free transitions using Hylleraas-style
correlated wave functions - an even more formidable computational
task than calculating the binding energy - and used them to explain
features in the solar continuous spectrum.
References:
S. Chandrasekhar, _Rev. Mod. Phys_. _16_, 301, 1944.
H. S. W. Massey, _Negative Ions_, Cambridge 1950.
H. A. Bethe and E. E. Salpeter, _Quantum Mechanics of One- and
Two-electron Atoms_, Plenum, 1977
------
Robert
Robert Parson
> Chandrasekhar then worked out the intensities
> of bound-bound and bound-free transitions ...
Gack, that should be "bound-free and free-free transitions". Not much
point to calculating bound-bound transitions in a system where there
is only one bound state to begin with. RP
Bob Casanova
J. Gans) wrote:
>Robert Grumbine (rm...@access5.digex.net) wrote:
>>In article <70g34c$4...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
>
>> A nice post which clarified just why there was a serious question
>>involved (a point which was bothering me, as I'd recently read Herzberg's
>>Atomic Spectra and Atomic Structure, and he seemed to think things were
>>pretty well managed).
>
>> My non-serious addition:
>
>>>The problem is that atoms with more than one electron
>>>(which, IIRC is most of them ;-)
>
>> Not by number! Most (ca. 92% by number, 75% by weight) of
>>the atoms in the universe only have 1 electron.
>
>>:-)
>
>>--
>>Robert Grumbine rm...@access.digex.net http://www.access.digex.net/~rmg3/
>>Sagredo (Galileo Galilei) "You present these recondite matters with too much
>>evidence and ease; this great facility makes them less appreciated than they
>>would be had they been presented in a more abstruse manner." Two New Sciences
>
>Well, there you go. So we can analytically solve Schroedinger's
>equation for 92% of the atoms in the universe. Since we can
>also do a *fine* _ab initio_ job on helium, which probably makes
>up a significant amount of the remaining 8%, other atoms are
>just "noise" and can be safely ignored. Quantum mechanics
>accurately explains 95+% of the atoms in the universe.
This reminds me of a description of the Solar system I encountered
(here, IIRC):
"One G0 star, 4 planets, and debris."
;-)
>
> ------ Paul J. Gans [ga...@panix.com]
>
>PS: Do I *need* a smiley?
Well, *I* usually add them as required; no telling who might be
reading this.
(Note followups, if any)
Bob C.
Reply to cas @ pop3.clark.net (without the spaces, of course)
"Men become civilized, not in proportion to their willingness
to believe, but in proportion to their readiness to doubt."
--H. L. Mencken
Paul J. Gans
> ------
> Robert
>
Thanks. I'm going to torture some of my theoretician-type
collegues with this.... heh heh heh..
Robert Grumbine
Bob Casanova <nos...@buzz.off> wrote:
>On 19 Oct 1998 18:15:45 -0400, in talk.origins, ga...@panix.com (Paul
>J. Gans) wrote:
>
>>Robert Grumbine (rm...@access5.digex.net) wrote:
>>> Not by number! Most (ca. 92% by number, 75% by weight) of
>>>the atoms in the universe only have 1 electron.
>>
>>>:-)
>>
>>Well, there you go. So we can analytically solve Schroedinger's
>>equation for 92% of the atoms in the universe. Since we can
>>also do a *fine* _ab initio_ job on helium, which probably makes
>>up a significant amount of the remaining 8%, other atoms are
>>just "noise" and can be safely ignored. Quantum mechanics
>>accurately explains 95+% of the atoms in the universe.
98+ % by number and weight.
>This reminds me of a description of the Solar system I encountered
>(here, IIRC):
>
>"One G0 star, 4 planets, and debris."
>
>;-)
>
>> ------ Paul J. Gans [ga...@panix.com]
>>
>>PS: Do I *need* a smiley?
>
>Well, *I* usually add them as required; no telling who might be
>reading this.
After seeing Robert Parson and Rich Trott taken for creationists
(or 'creationist sympathizers'), I wasn't taking any chances.
tpham...@my-dejanews.com
By coincidence I have just covered the variational method and a
problem in "Quantum Mechanics in Chemistry" by Jack Simons and
Jeff Nichols also points this out. They don't show the calculations
with the trial functions using explicit r12 dependence
(1 and 2 refer to electrons 1 and 2) - wimpy chemists!
MJR
Take that back! Or it will be naked sabres at dawn.
Seriously, Jack Simons is one of the finest human beings on the face
of this planet. I learned my introductory electronic structure theory
from the first draft of the text, back when Jeff Nichols was still at
Utah ( and when IBM still saw the wisdom of supporting research units in
close collaboration with universities ). If Jack doesn't see fit to
treat explicitly correlated wavefunctions in his text, it's because they
are *ahem* not relevant to the kind of manly calculations we do on real
chemical systems :-)
Mogens
Paul J. Gans
Hmm. All of this arose because of the article in the
Scientific American. One of the authors was here posting
some rather strange statements about quantum mechanics.
Some of us didn't agree with him (electron correlation
or not).
Wonder where he went? He seemed to have some serious
misconceptions. I may be wrong about this, but he'd
have to point out where we were wrong.
Robert Parson
> Seriously, Jack Simons is one of the finest human beings on the face
>of this planet.
Gotta agree; after all he invited me to give a talk at the 1996 ACTC :-)
Speaking of which, I do hope everyone in this sub-sub-thread is
planning to come to the 1999 ACTC in Boulder, CO. Maybe I can
convince the organizers to schedule a session on the applications of
Intelligent Design Theory to Quantum Mechanics (or maybe
applications of Quantum Mechanics to Intelligent Design Theory.)
After which we can head into the Arapahoe National Forest for
some Large-Mouthed Bass fishing (though out here you're more likely
to get trout.)
> I learned my introductory electronic structure theory from the first
> draft of the text, back when Jeff Nichols was still at Utah
It's a good book, which does a lot to demystify that ghastly jungle
of acronyms that has permeated the field. BTW, Szabo and Ostlund's
_Modern Quantum Chemistry_, another great book (at a higher level than
Simons and Nichols, though) is now available in a cheap Dover paperback
edition.
------
Robert
MJR
>
> In article <362EA3...@kemi.aau.dk>, MJR <johan...@kemi.aau.dk> wrote:
>
> > Seriously, Jack Simons is one of the finest human beings on the face
> >of this planet.
>
> Gotta agree; after all he invited me to give a talk at the 1996 ACTC :-)
> Speaking of which, I do hope everyone in this sub-sub-thread is
> planning to come to the 1999 ACTC in Boulder, CO. Maybe I can
> convince the organizers to schedule a session on the applications of
> Intelligent Design Theory to Quantum Mechanics (or maybe
> applications of Quantum Mechanics to Intelligent Design Theory.)
> After which we can head into the Arapahoe National Forest for
> some Large-Mouthed Bass fishing (though out here you're more likely
> to get trout.)
Cool! We must have met. I was one of the monkey boys, scampering
about in the red, white, & blue bowling shirts to run slide projectors
and to lean on the hotel staff for more beer. I must confess that I
don't remember you, as I was finishing my thesis in the usual hysterical
daze.
It's none too soon to start bribing the travel committee for a trip
to Colorado.
>
> > I learned my introductory electronic structure theory from the first
> > draft of the text, back when Jeff Nichols was still at Utah
>
> It's a good book, which does a lot to demystify that ghastly jungle
> of acronyms that has permeated the field. BTW, Szabo and Ostlund's
> _Modern Quantum Chemistry_, another great book (at a higher level than
> Simons and Nichols, though) is now available in a cheap Dover paperback
> edition.
>
Complementarity in action: Szabo and Ostlund show you how to do it,
Simons and Nichols teach you when to do what.
> ------
> Robert
tpham...@my-dejanews.com
rpa...@spotNO.SPAMColorado.edu (Robert Parson) wrote:
> In article <362EA3...@kemi.aau.dk>, MJR <johan...@kemi.aau.dk> wrote:
> > Seriously, Jack Simons is one of the finest human beings on the face
> >of this planet.
> Gotta agree; after all he invited me to give a talk at the 1996 ACTC :-)
> Speaking of which, I do hope everyone in this sub-sub-thread is
> planning to come to the 1999 ACTC in Boulder, CO. Maybe I can
> convince the organizers to schedule a session on the applications of
> Intelligent Design Theory to Quantum Mechanics (or maybe
> applications of Quantum Mechanics to Intelligent Design Theory.)
> After which we can head into the Arapahoe National Forest for
> some Large-Mouthed Bass fishing (though out here you're more likely
> to get trout.)
Sounds like a Howler-fest opportunity! I plan to be there, but
like one of them wimpy chemists, probably not with any wave functions
with explicit r12 dependence in the basis set (you never know!)
> > I learned my introductory electronic structure theory from the first
> > draft of the text, back when Jeff Nichols was still at Utah
> It's a good book, which does a lot to demystify that ghastly jungle
> of acronyms that has permeated the field. BTW, Szabo and Ostlund's
> _Modern Quantum Chemistry_, another great book (at a higher level than
> Simons and Nichols, though) is now available in a cheap Dover paperback
> edition.
I second those recommendations, even though they don't tell how all
that information in chemistry gets crammed into the time-independent
Schrodinger equation. That just shows the need for the cutting-edge
symposia proposed by Robert.
Robert Parson
> Cool! We must have met. I was one of the monkey boys, scampering
>about in the red, white, & blue bowling shirts to run slide projectors
>and to lean on the hotel staff for more beer. I must confess that I
>don't remember you, as I was finishing my thesis in the usual hysterical
>daze.
Well, my talk was at 8:00 AM on Friday morning, the last day of the
conference, and we'd just had that glorious banquet the preceding night,
so attendance was down by a little (and attention spans by a lot more.)
Which is just as well, since I am currently working on a paper which
demonstrates conclusively that one of the major arguments in that talk
was completely wrong.
------
Robert
Matt Silberstein
rpa...@spotNO.SPAMColorado.edu (Robert Parson):
Thus works science. Not only do scientists work hard to refute other
scientists, they work to refute themselves. Take that, creationists.
Matt Silberstein
-------------------------------------------------------
Ain't there one damn song that can make me
break down and cry.
David Bowie
Rex Nielson
> MJR (johan...@kemi.aau.dk) wrote:
> >tpham...@my-dejanews.com wrote:
> >>
> >> In article <70ihl5$a...@peabody.colorado.edu>,
> > Seriously, Jack Simons is one of the finest human beings on the face
> >of this planet. I learned my introductory electronic structure theory
> >from the first draft of the text, back when Jeff Nichols was still at
> >Utah ( and when IBM still saw the wisdom of supporting research units in
> >close collaboration with universities ). If Jack doesn't see fit to
> >treat explicitly correlated wavefunctions in his text, it's because they
> >are *ahem* not relevant to the kind of manly calculations we do on real
> >chemical systems :-)
>
>
>
> Hmm. All of this arose because of the article in the
> Scientific American. One of the authors was here posting
> some rather strange statements about quantum mechanics.
> Some of us didn't agree with him (electron correlation
> or not).
>
> Wonder where he went? He seemed to have some serious
> misconceptions. I may be wrong about this, but he'd
> have to point out where we were wrong.
>
> ------ Paul J. Gans [ga...@panix.com]
It seemed to me like you guys both agreed that QM can't describe the whole
periodic table. What's the serious misconception you think he may have?
Is it that he brought up a "wrong question to ask"?
That descriptions for elements heavier then helium are a "convenient
fiction" is news to me. Perhaps you should be more careful about what you
say in front of impressionable ears. What do you think about Denton's
proposal, from his new book, that the naturally occurring elements are so
supremely fit for life, that they are literally "meant" for life? It would
seem a little presumptuous to reject off hand the argument that the
elements might be "for" something, in light of your confession that science
can't even tell us exactly what they are.
-RN
Paul J. Gans
Say what? I was under the strange impression that I argued
no such thing. It is true that _ab initio_ calculations on
heavy elements are difficult and that toward the end of the
periodic chart one needs to use a relativistic version of
quantum mechanics, but "convenient fiction?"
Our author had claimed that there was tomfoolery in chosing
special "basis sets" for doing the calculations on copper,
though he admitted that those gave answers that agreed with
experiment. At that point I noted that he seemed not to
understand what "basis sets" are and why the choice of
basis set only affected the efficiency of the computation
and not, if enough terms are taken, the result.
N.B.: A basis set is something roughly analogous to a
coordinate system. For instance one can work a problem
in cartesian coordinates (x,y,z), or spherical coordinates
(r,theta,phi), or cylindrical coordinates (r,theta,z), or
in one of a large number of other systems. The physical
results will be the same, though the computations may be
*much* easier in one system than in another. For instance,
if one wanted to deal with a problem involving spheres,
a spherical coordinate system might well be the one of
choice. In that same manner, quantum mechanicians choose
basis sets for their ability to mimic the known features
of the problem. It just makes the work easier.
>What do you think about Denton's
>proposal, from his new book, that the naturally occurring elements are so
>supremely fit for life, that they are literally "meant" for life? It would
>seem a little presumptuous to reject off hand the argument that the
>elements might be "for" something, in light of your confession that science
>can't even tell us exactly what they are.
>-RN
I made no such confession. Or if I did, I pleasd temporary
insanity. I've not read Denton, but the argument is *obviously*
circular. Before the elements can be characterized, there must
be "intelligent" life present to characterize them. Since then
life exists, of course the elements are "fit" for life.
But of course they are not "supremely" fit. Too many of them
are dangerous and/or poisonous.
And, I can't help deducing that Denton, if the remarks above
accurately mirror his opinion, doesn't know much about either
chemistry or the elements.
Robert Parson
>>> Hmm. All of this arose because of the article in the
>>> Scientific American. One of the authors was here posting
>>> some rather strange statements about quantum mechanics.
>>> Some of us didn't agree with him (electron correlation
>>> or not).
>>>
>>> Wonder where he went? He seemed to have some serious
>>> misconceptions. I may be wrong about this, but he'd
>>> have to point out where we were wrong.
Well I, for one, think he has a point, although he did not express
it very well. Let me point out that I am not an ab initio
quantum chemist myself, I'm a dynamicist who works with ab initio
results so I hope TPH and MJR will set me straight when I get
the details wrong as I probably will.
>Our author had claimed that there was tomfoolery in chosing
>special "basis sets" for doing the calculations on copper,
>though he admitted that those gave answers that agreed with
>experiment. At that point I noted that he seemed not to
>understand what "basis sets" are and why the choice of
>basis set only affected the efficiency of the computation
>and not, if enough terms are taken, the result.
But in practice "enough terms" are _never_ taken (except in
something really simple like a two-electron system). The
Hilbert space of a many-electron atom or molecule is
very, very, very big. There is no hope of covering
it with enough basis functions so that everything will
converge.
>choice. In that same manner, quantum mechanicians choose
>basis sets for their ability to mimic the known features
>of the problem. It just makes the work easier.
No, it makes the work _possible_. Quantum chemistry methods
are chosen, at least in part, according to their track record
for giving correct results in similar problems. There are
heuristic arguments for these decisions: for example, if you
want to treat a negative ion you should have diffuse functions
in your atomic basis set because anion wave functions tend to
spread out farther; if lone pairs are important to you, you
should use a coupled-cluster method rather than MP perturbation
theory because CC builds in pair correlations, etc. But these
are heuristics, insights derived from physical intuition, chemical
intuition, and the results of comparing theory with experiment
on simpler problems. If you ignore the wisdom of the professionals,
you don't get a calculation that converges slowly, you get a
calculation that converges quite nicely to the wrong result.
This is still more true in density functional theory: the
exchange-correlation functionals that are in common use now seem
to be based upon about 30% physical intuition, 40% accumulated
experience with older functionals that didn't work, and 30% black magic.
So I do think it is appropriate to ask whether ab initio quantum
chemistry calculations really derive their results from "first
principles." Let me emphasize that I'm not singling out quantum
chemistry in this respect; much the same is true of all cutting-
edge research in theoretical physics and chemistry. Rigorous,
a priori error estimates are rarely available for the most
exciting experimental problems. If we restricted ourselves to
problems where we could state error bounds, we'd probably be stuck
somewhere around Beryllium.
This is why questions of "prediction" in science are not simple.
Ab initio quantum chemistry has reached the stage where disagreements
between theory and experiment are not always resolved in favor of
the experimentalist. (Yes, Tracy, I know you want to say something
about methylene here. :-) ). That pragmatic criterion is what, for
me, demonstrates the success of the field.
------
Robert
Rex Nielson
> Say what? I was under the strange impression that I argued
> no such thing. It is true that _ab initio_ calculations on
> heavy elements are difficult and that toward the end of the
> periodic chart one needs to use a relativistic version of
> quantum mechanics, but "convenient fiction?"
See your post on the 19th in this same thread.
[quote]
Can we explain this via quantum mechanics? As Eric
Scerri has pointed out, no, we cannot.
It is something like asking "how does the stork
manage to carry a 12 pound baby through the air?"
The answer is that a stork does no such thing. And
the answer for copper is that there really isn't any
such thing as 4s or 3d electrons. They are a
convenient fiction. [sic] More technically, there aren't
any "good" quantum numbers for such things in
multi-electron atoms.
[end quote]
This post was fascinating to me because I had assumed that a description of
the periodic table would be among QM's successes. In the post you seemed
to indicate that the inability to explain the configuration of the heavier
elements was more then the inconvenience of the calculation that would be
required but because "there are no sets of "pure" electronic states for
multi-electron atoms". Is the common assumption that chemistry reduces to
physics, just that, an assumption? That would seem to be the implication
if the structure of the heavier elements is, for physicists, mere
conjecture.
>
> Our author had claimed that there was tomfoolery in chosing
> special "basis sets" for doing the calculations on copper,
> though he admitted that those gave answers that agreed with
> experiment. At that point I noted that he seemed not to
> understand what "basis sets" are and why the choice of
> basis set only affected the efficiency of the computation
> and not, if enough terms are taken, the result.
>
> N.B.: A basis set is something roughly analogous to a
> coordinate system. For instance one can work a problem
> in cartesian coordinates (x,y,z), or spherical coordinates
> (r,theta,phi), or cylindrical coordinates (r,theta,z), or
> in one of a large number of other systems. The physical
> results will be the same, though the computations may be
> *much* easier in one system than in another. For instance,
> if one wanted to deal with a problem involving spheres,
> a spherical coordinate system might well be the one of
> choice. In that same manner, quantum mechanicians choose
> basis sets for their ability to mimic the known features
> of the problem. It just makes the work easier.
Whatever coordinate system is used wouldn't it rely on the assumption that
there is something in space to designate? Doesn't QM rest on the
conjecture that when matter behaves like a wave, it's actually an illusion
created by particles? What if particles are an illusion created by waves?
Physicist Mendel Sachs argues that the peace between QM and relativity is a
false peace and that with the application of relativity to the micro-level
one would end up with a "wave monism" where waves instead of particles are
seen to constitute matter. It does seem strange to me that the
deterministic continuum of relativity would cut off at a certain level of
smallness and arbitrary quantum activity would commence.
I'm supremely unqualified to comment on the plausibility of theories of
physics by interpretation of experiment or in terms mathematical proofs,
but I think I can see a curious pattern at work. A few biologists today
ridicule neoDarwinism for its invitation to describe evolution as a story
about the adventures of the genes. I think the inspiration for this comes
ultimately from mechanistic and atomistic assumptions that run deep in
science and are not held by biologists alone. Perhaps future physicists
will make fun of their predecessors for having described material phenomena
as 'the adventures of the particles'.
> >What do you think about Denton's
> >proposal, from his new book, that the naturally occurring elements are so
> >supremely fit for life, that they are literally "meant" for life? It would
> >seem a little presumptuous to reject off hand the argument that the
> >elements might be "for" something, in light of your confession that science
> >can't even tell us exactly what they are.
>
> >-RN
>
> I made no such confession. Or if I did, I pleasd temporary
> insanity. I've not read Denton, but the argument is *obviously*
> circular. Before the elements can be characterized, there must
> be "intelligent" life present to characterize them. Since then
> life exists, of course the elements are "fit" for life.
>
> But of course they are not "supremely" fit. Too many of them
> are dangerous and/or poisonous.
>
> And, I can't help deducing that Denton, if the remarks above
> accurately mirror his opinion, doesn't know much about either
> chemistry or the elements.
>
> ----- Paul J. Gans [ga...@panix.com]
Uranium may be bad for our personal health, but we wouldn't have our
biosphere if it wasn't around to fuel tectonic activity.
-RN
Paul J. Gans
>>>> Hmm. All of this arose because of the article in the
>>>> Scientific American. One of the authors was here posting
>>>> some rather strange statements about quantum mechanics.
>>>> Some of us didn't agree with him (electron correlation
>>>> or not).
>>>>
>>>> Wonder where he went? He seemed to have some serious
>>>> misconceptions. I may be wrong about this, but he'd
>>>> have to point out where we were wrong.
> Well I, for one, think he has a point, although he did not express
> it very well. Let me point out that I am not an ab initio
> quantum chemist myself, I'm a dynamicist who works with ab initio
> results so I hope TPH and MJR will set me straight when I get
> the details wrong as I probably will.
>>Our author had claimed that there was tomfoolery in chosing
>>special "basis sets" for doing the calculations on copper,
>>though he admitted that those gave answers that agreed with
>>experiment. At that point I noted that he seemed not to
>>understand what "basis sets" are and why the choice of
>>basis set only affected the efficiency of the computation
>>and not, if enough terms are taken, the result.
> But in practice "enough terms" are _never_ taken (except in
> something really simple like a two-electron system). The
> Hilbert space of a many-electron atom or molecule is
> very, very, very big. There is no hope of covering
> it with enough basis functions so that everything will
> converge.
>>choice. In that same manner, quantum mechanicians choose
>>basis sets for their ability to mimic the known features
>>of the problem. It just makes the work easier.
> No, it makes the work _possible_. Quantum chemistry methods
> ------
> Robert
Can't argue with Robert here, though I think he understands
that I was trying very hard to keep things on a level that
can be comprehended here.
The point is: could we, if we wanted to, do a decent
quantum-mechanical computation on copper that would give
the "right" answers. This may already have been done for
all I know. I did not do a library search (I just asked
the first handy quantum mechanician who chanced to walk
by my door -- one Mark Tuckerman who is in somewhat the
same business that Robert is in.)
The answer is that we could do such a computation. It
would be rather expensive (the space is *rather* large)
but we could do it, and I'm willing to bet a few beers
that the result would be quite satisfactory.
Such problems are not what folks are presently interested
in. Robert's dynamics using more or less exact potential
surfaces computed via density functional methods is popular
now, as are several other things.
But we aren't going to go there, are we?
Paul J. Gans
I was afraid of this. I'm going to have to give you a
very unsatisfactory answer, I'm afraid.
In general chemistry courses we teach students that
electrons in an atom live in neat little arrangements
known as "orbitals". Thus we speak of 3d orbitals
and 5s orbitals, etc.
Alas, this is not completely true. It is certainly true
of hydrogen with only one electron, but for any other atom
the interaction between electrons keep things from being
quite so neat. What is true is that the energies of electrons
in multi-electron atoms are quantized, but again not in
the manner usually thought. For example, the minimum
energy required to remove an electron from a copper atom
is usually called the "first ionization energy" and it is
a fixed (and known) quantity. However, the identity of
the electron removed is unknown and unknowable, as all
electrons are identical. If we take the resulting ion
and remove another electron using the least possible
energy, the resulting fixed (and usually known quantity)
is called the "second ionization energy".
However, if we go back to the original atom and shine
photons having energy corresponding to the second ionization
energy on that atom, we may or may not remove and electron.
The energy levels in the atom are *different* than those
in the ion. In other words, the energy levels of the
copper atom are *not* fixed for all time, but depend on
the number of electrons in that atom at a particular
moment in time.
The implication of this, and similar experiments, is that
electrons do not live in convenient boxes with neat labels
such as "4p".
Having said this in (too many) words, what I was talking
about in the material you quoted was this: the original
auther may (it was never clear to me) have been asking
if quantum mechanics can calculate the energy of a "3s"
electron in copper. My answer is no, NOT because
quantum mechanics doesn't work, but because the notion
of a 3s orbital in copper is wrong.
>> Our author had claimed that there was tomfoolery in chosing
>> special "basis sets" for doing the calculations on copper,
>> though he admitted that those gave answers that agreed with
>> experiment. At that point I noted that he seemed not to
>> understand what "basis sets" are and why the choice of
>> basis set only affected the efficiency of the computation
>> and not, if enough terms are taken, the result.
>>
>> N.B.: A basis set is something roughly analogous to a
>> coordinate system. For instance one can work a problem
>> in cartesian coordinates (x,y,z), or spherical coordinates
>> (r,theta,phi), or cylindrical coordinates (r,theta,z), or
>> in one of a large number of other systems. The physical
>> results will be the same, though the computations may be
>> *much* easier in one system than in another. For instance,
>> if one wanted to deal with a problem involving spheres,
>> a spherical coordinate system might well be the one of
>> choice. In that same manner, quantum mechanicians choose
>> basis sets for their ability to mimic the known features
>> of the problem. It just makes the work easier.
>Whatever coordinate system is used wouldn't it rely on the assumption that
>there is something in space to designate? Doesn't QM rest on the
>conjecture that when matter behaves like a wave, it's actually an illusion
>created by particles? What if particles are an illusion created by waves?
This gets into deep merde and I don't really want to go there.
Quantum mechanics, if viewed from a classical perspective, is
filled with paradoxes. The short answer is that electrons
(and other elementary entities) are just entities. If you
experimentally treat them like particles they (usually) give
you particle-like behavior. And if you experimentally treat
them like waves they (usually) give you wave-like behavior.
That does not mean that they are either particles or waves.
They are entities.
Once folks kicked the classical habit they began thinking of
these entities in terms of what we could know about them
and stopped worrying about what we cannot know.
>Physicist Mendel Sachs argues that the peace between QM and relativity is a
>false peace and that with the application of relativity to the micro-level
>one would end up with a "wave monism" where waves instead of particles are
>seen to constitute matter. It does seem strange to me that the
>deterministic continuum of relativity would cut off at a certain level of
>smallness and arbitrary quantum activity would commence.
I'm not a relativistic quantum mechanician. But I'm fairly
certain that there is no arbitrary cut-off. If we look at
the similar situation between QM and classical physics we
see that the transition is rather smooth, although there
are criteria as to when the transition will take place.
>I'm supremely unqualified to comment on the plausibility of theories of
>physics by interpretation of experiment or in terms mathematical proofs,
>but I think I can see a curious pattern at work. A few biologists today
>ridicule neoDarwinism for its invitation to describe evolution as a story
>about the adventures of the genes. I think the inspiration for this comes
>ultimately from mechanistic and atomistic assumptions that run deep in
>science and are not held by biologists alone. Perhaps future physicists
>will make fun of their predecessors for having described material phenomena
>as 'the adventures of the particles'.
Well, I'm unqualified to comment on this.
>
>> >What do you think about Denton's
>> >proposal, from his new book, that the naturally occurring elements are so
>> >supremely fit for life, that they are literally "meant" for life? It would
>> >seem a little presumptuous to reject off hand the argument that the
>> >elements might be "for" something, in light of your confession that science
>> >can't even tell us exactly what they are.
>>
>> >-RN
>>
>> I made no such confession. Or if I did, I pleasd temporary
>> insanity. I've not read Denton, but the argument is *obviously*
>> circular. Before the elements can be characterized, there must
>> be "intelligent" life present to characterize them. Since then
>> life exists, of course the elements are "fit" for life.
>>
>> But of course they are not "supremely" fit. Too many of them
>> are dangerous and/or poisonous.
>>
>> And, I can't help deducing that Denton, if the remarks above
>> accurately mirror his opinion, doesn't know much about either
>> chemistry or the elements.
>>
>> ----- Paul J. Gans [ga...@panix.com]
>Uranium may be bad for our personal health, but we wouldn't have our
>biosphere if it wasn't around to fuel tectonic activity.
It doesn't fuel tectonic activity. At least not alone. And
if you claim that "supremely fit" elements have both a good
side and a bad side, then you are warping what I'd understand
by "supremely fit".
There exist elements absolutely useless to life. Neon is one.
How does it come to be "supremely fit"?
wf...@enter.netxx
Parson) wrote:
> This is why questions of "prediction" in science are not simple.
> Ab initio quantum chemistry has reached the stage where disagreements
> between theory and experiment are not always resolved in favor of
> the experimentalist. (Yes, Tracy, I know you want to say something
> about methylene here. :-) ). That pragmatic criterion is what, for
> me, demonstrates the success of the field.
>
a good example of this is the ongoing dispute between fritz schaeffer
and f. a. cotton over whether triple Ga-Ga bonds exist. both cite
theory and experimental evidence to say they do/dont...interesting
Rex Nielson
Paul Gans wrote:
>>Whatever coordinate system is used wouldn't it rely on the assumptionthat
>>there is something in space to designate? Doesn't QM rest on the
>>conjecture that when matter behaves like a wave, it's actually an illusion
>>created by particles? What if particles are an illusion created by waves?
>This gets into deep merde and I don't really want to go there.
>Quantum mechanics, if viewed from a classical perspective, is
>filled with paradoxes. The short answer is that electrons
>(and other elementary entities) are just entities. If you
>experimentally treat them like particles they (usually) give
>you particle-like behavior. And if you experimentally treat
>them like waves they (usually) give you wave-like behavior.
>That does not mean that they are either particles or waves.
>They are entities.
>Once folks kicked the classical habit they began thinking of
>these entities in terms of what we could know about them
>and stopped worrying about what we cannot know.
>>Physicist Mendel Sachs argues that the peace between QM and relativity a
>>false peace and that with the application of relativity to the micro-level
>>one would end up with "wave monism" where waves instead of particles
>>seen to constitute matter. It does seem strange to me that the
>>deterministic continuum of relativity would cut off at a certain level of
>>smallness and arbitrary quantum activity would commence.
>I'm not a relativistic quantum mechanician. But I'm fairly
>certain that there is no arbitrary cut-off. If we look at
>the similar situation between QM and classical physics we
>see that the transition is rather smooth, although there
>are criteria as to when the transition will take place.
>>I'm supremely unqualified to comment on the plausibility of theories of
>>physics by interpretation of experiment or in terms mathematical proofs,
>>but I think I can see a curious pattern at work. A few biologists today
>>ridicule neoDarwinism for its invitation to describe evolution as a story
>>about the adventures of the genes. I think the inspiration for this comes
>>ultimately from mechanistic and atomistic assumptions that run deep in
>>science and are not held by biologists alone. Perhaps future physicists
>>will make fun of their predecessors for having described material >>phenomena
>>as 'the adventures of the particles'.
>Well, I'm unqualified to comment on this.
>>Uranium may be bad for our personal health, but we wouldn't have our
>>biosphere if it wasn't around to fuel tectonic activity.
>It doesn't fuel tectonic activity. At least not alone. And
>if you claim that "supremely fit" elements have both a good
>side and a bad side, then you are warping what I'd understand
>by "supremely fit".
>There exist elements absolutely useless to life. Neon is one.
>How does it come to be "supremely fit"?
----- Paul J. Gans [ga...@panix.com]
Well okay, thorium to. But you'll have to concede that the heating of the
earth and subsequent tectonic activity is the result of the radioactive
decay of the heavy elements, unless you want to follow Lord Kelvin and say
the earth considerably younger then is thought today.
"Supremely fit" is a term used by Denton in his book _Nature's Design_. He
admits that the cosmos cannot be shown to be "uniquely fit" for
life-as-know-it because it's impossible to state conclusively that
life-as-we-know-it is the only way life could be. Everyone can agree that
the universe is at least fit enough for life for life to exist, but Denton
points to the large series of natural phenomena that display singular
qualities necessary for life and argues that the fitness of the cosmos is
too extensive to be a meaningless coincidence:
[quote]
If these various constituents---water, carbon dioxide, carbonic acid, the
DNA helix, proteins, phosphates, sugars, lipids, the carbon atom, the
oxygen atom, the transitional metal atoms and the other metal atoms from
groups 1 and 2 of the periodic table, sodium, potassium, calcium, and
magnesium---did not possess precisely those chemical and physical
properties they exhibit in an aqueous solution ranging in temperature of 0
degrees Celsius and about 75 degrees Celsius, self-replicating carbon-based
chemical machines would be impossible.
[end quote---p. 381 _ND_]
As for the noble gases, Denton suggests they are necessary by-products of
the atom-building rules. Perhaps seeming aberrations can be justified in
the context of a larger scheme. One can, of course, justify anything with
this kind of reasoning and with recourse to the explanation of "higher law"
there is potential for abuse (e.g. God made neon for lightbulbs). But if
evidence is clear that the universe may be governed by higher laws,
students of nature are obliged to pursue and discover them, even if this
contradicts strongly held conventions.
-RN
wf...@enter.netxx
wrote:
but Denton
>points to the large series of natural phenomena that display singular
>qualities necessary for life and argues that the fitness of the cosmos is
>too extensive to be a meaningless coincidence:
which proves exactly nothing at all. we dont know that this is the
only universe; lee smolin, andrei linde, and others have argued that
its not. and those arguments are based on concepts in science, not
concepts in magic.
design theorists say that absence of evidence is evidence of absence;
since we dont know all the answers, making up answers is rational.
that's not science, its guesswork.
But if
>evidence is clear that the universe may be governed by higher laws,
>students of nature are obliged to pursue and discover them, even if this
>contradicts strongly held conventions.
>
>-RN
>
except that design theorists have never articulated any rules at all
about their designer, or how he works. their theory is sterile.
MJR
>
> In article <70onua$7...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
>
> >>> Hmm. All of this arose because of the article in the
> >>> Scientific American. One of the authors was here posting
> >>> some rather strange statements about quantum mechanics.
> >>> Some of us didn't agree with him (electron correlation
> >>> or not).
> >>>
> >>> Wonder where he went? He seemed to have some serious
> >>> misconceptions. I may be wrong about this, but he'd
> >>> have to point out where we were wrong.
>
> it very well. Let me point out that I am not an ab initio
> quantum chemist myself, I'm a dynamicist who works with ab initio
> results so I hope TPH and MJR will set me straight when I get
> the details wrong as I probably will.
I have found the comments by Paul and Robert to be quite correct,
gratifyingly so compared to the pointless drivel so often spewed out
upon this newsgroup.
>
> >Our author had claimed that there was tomfoolery in chosing
> >special "basis sets" for doing the calculations on copper,
> >though he admitted that those gave answers that agreed with
> >experiment. At that point I noted that he seemed not to
> >understand what "basis sets" are and why the choice of
> >basis set only affected the efficiency of the computation
> >and not, if enough terms are taken, the result.
>
> But in practice "enough terms" are _never_ taken (except in
> something really simple like a two-electron system). The
> Hilbert space of a many-electron atom or molecule is
> very, very, very big. There is no hope of covering
> it with enough basis functions so that everything will
> converge.
>
This point is so fundamental that it ought to be drilled into every
first year student in computational science. Computers are stupid: once
an input deck is submitted the machine will merrily follow the algorithm
until the STOP signal is detected. The user is the one who is supposed
to be smart.
The problem is quite severe in chemistry, so much so that I have
something of a standard speech I give every time an experimentalist
comes to me after reading an article about the marvel advances in
chemistry software; it runs along the lines of "what you are suggesting
is *way* too difficult to accomplish in one lifetime, and the numbers I
*can* compute are of no obvious relevence to your chemical problem. So,
let's both think about this real hard and see if we can at least shine
some new light on the problem and inspire a more clever line of
investigation".
Your comments on convergence might be misleading to the non expert;
most people work with the integral/algebraic approximation to the
Schroedinger equation rather than, strictly speaking, solving the
differential equation form as this practically guarantees a numerically
stable solution-which may of course be numerically precise garbage. The
larger point is completely valid: as the basis is expanded the computed
value converges towards the complete limit and the investigator has to
think in terms of "what question am I trying to answer, does the method
I use incorporate the essential physics, and just how much accuracy to I
need?"
As Paul observes, some 'problems' are in a sense completely fake.
The whole idea of a 3d electron arises from the mean field approxima-
tion; somewhat miraculously, this approach has been developed into a
simple and powerful theory of chemical structure and bonding that
researchers can manipulate within their heads to obtain important
insights into the properties of matter. For this reason, not because
electron shells are "real" in any formal sense, the standard theory of
electronic structure has become one of the main pillars in the chemical
edifice.
> >choice. In that same manner, quantum mechanicians choose
> >basis sets for their ability to mimic the known features
> >of the problem. It just makes the work easier.
>
"Easier" and "possible" are in a deep sense synonymous in quantum
mechanics. One of the striking features of QM is how easily people,
after a modest introduction to the field, can propse impossibly
difficult problems whose solutions would be of immense value. The art
lies in posing problems that are simultaneously solvable and useful, and
for that we rely-as must workers in all fields-on the wisdom handed down
from those who came before. "If I have not seen so far as other men,
'tis because giants were standing on my shoulders".
>
> This is why questions of "prediction" in science are not simple.
> Ab initio quantum chemistry has reached the stage where disagreements
> between theory and experiment are not always resolved in favor of
> the experimentalist. (Yes, Tracy, I know you want to say something
> about methylene here. :-) ). That pragmatic criterion is what, for
> me, demonstrates the success of the field.
>
> ------
> Robert
I seem to be particularly garrulous this morning, so at the risk of
taking this deeper into the domain of soc.culture.comp-chem, I would
like to ask what the smart guys think about Pople getting the Nobel
Prize. Granted, without the Gaussian codes no one outside an incestuous
clique would have ever heard of electronic structure theory, but Pople
has already been rewarded for that with his stock in Gaussian Inc. As a
colleague, who shall for obvious reasons remain anonymous, remarked:
Pople never had an original idea in his life.
For a man whose contribution to the art is comparable to that of
Walter Kohn in Density Functional Theory, I would argue that Enrico
Clemmenti or Per-Olav Lowdin should have been selected. As a torpedo
for the Scandnavian Mafia, I of course would have given the nod to Per-
Olav; not only did he contribute a series of seminal papers, but by
winkling funds from the US Air Force and training a raft of brilliant
graduate students he did much to establish the modern school of
electronic structure theory upon which all of our currect success is
based.
End of sermon.
tpham...@my-dejanews.com
The orbital concept is based on the Hartree-Fock approximation,
usually.
> >>Whatever coordinate system is used wouldn't it rely on the assumptionthat
> >>there is something in space to designate?
The coordinate system is irrelevant.
> >>Doesn't QM rest on the
> >>conjecture that when matter behaves like a wave, it's actually an illusion
> >>created by particles? What if particles are an illusion created by waves?
Short answer to the questions: No. and How would we know?
> >This gets into deep merde and I don't really want to go there.
> >Quantum mechanics, if viewed from a classical perspective, is
> >filled with paradoxes. The short answer is that electrons
> >(and other elementary entities) are just entities. If you
> >experimentally treat them like particles they (usually) give
> >you particle-like behavior. And if you experimentally treat
> >them like waves they (usually) give you wave-like behavior.
> >That does not mean that they are either particles or waves.
> >They are entities.
> >Once folks kicked the classical habit they began thinking of
> >these entities in terms of what we could know about them
> >and stopped worrying about what we cannot know.
That is, we *can't* know everything.
> >>Physicist Mendel Sachs argues that the peace between QM and relativity a
> >>false peace and that with the application of relativity to the micro-level
> >>one would end up with "wave monism" where waves instead of particles
> >>seen to constitute matter. It does seem strange to me that the
> >>deterministic continuum of relativity would cut off at a certain level of
> >>smallness and arbitrary quantum activity would commence.
Particles and waves are *concepts*. That is the whole point to
"particle-wave" duality. Just because a particle can be modelled
by a wave packet does not mean it *is* a wave packet. It is all a
matter of which concept is more useful in which circumstance.
> >I'm not a relativistic quantum mechanician. But I'm fairly
> >certain that there is no arbitrary cut-off. If we look at
> >the similar situation between QM and classical physics we
> >see that the transition is rather smooth, although there
> >are criteria as to when the transition will take place.
> >>I'm supremely unqualified to comment on the plausibility of theories of
> >>physics by interpretation of experiment or in terms mathematical proofs,
> >>but I think I can see a curious pattern at work. A few biologists today
> >>ridicule neoDarwinism for its invitation to describe evolution as a story
> >>about the adventures of the genes. I think the inspiration for this comes
> >>ultimately from mechanistic and atomistic assumptions that run deep in
> >>science and are not held by biologists alone. Perhaps future physicists
> >>will make fun of their predecessors for having described material
>>phenomena
> >>as 'the adventures of the particles'.
I don't see the pattern you are referring to. All I see are
refinements of or additions to still good theories.
[snip?]
> >It doesn't fuel tectonic activity. At least not alone. And
> >if you claim that "supremely fit" elements have both a good
> >side and a bad side, then you are warping what I'd understand
> >by "supremely fit".
>
> >There exist elements absolutely useless to life. Neon is one.
> >How does it come to be "supremely fit"?
> Well okay, thorium to. But you'll have to concede that the heating of the
> earth and subsequent tectonic activity is the result of the radioactive
> decay of the heavy elements, unless you want to follow Lord Kelvin and say
> the earth considerably younger then is thought today.
But is tectonic activity necessary for life?
> "Supremely fit" is a term used by Denton in his book _Nature's Design_. He
[snip]
> As for the noble gases, Denton suggests they are necessary by-products of
> the atom-building rules. Perhaps seeming aberrations can be justified in
> the context of a larger scheme. One can, of course, justify anything with
> this kind of reasoning and with recourse to the explanation of "higher law"
> there is potential for abuse (e.g. God made neon for lightbulbs). But if
> evidence is clear that the universe may be governed by higher laws,
> students of nature are obliged to pursue and discover them, even if this
> contradicts strongly held conventions.
Until we could be absoulutely certain about what impact a slight change
in some fundamental aspect of the universe (Or if they are even
variable, or even other universes possible) then "how nicely things seem
to fit together" only supports various idle speculations.
-----------------------------------------------------------------------
Tracy P. Hamilton |O Beer! O Hodgson, Guinness, Alsopp, Bass!
Building Manager, Alco Hall |Names that should be on every infants' tongue!
University of Ediacara | - Charles Stuart Calverly
-----------------------------------------------------------------------
nothing about the universe.
tpham...@my-dejanews.com
rpa...@spotNO.SPAMColorado.edu (Robert Parson) wrote:
> >>> Hmm. All of this arose because of the article in the
> >>> Scientific American. One of the authors was here posting
> >>> some rather strange statements about quantum mechanics.
> >>> Some of us didn't agree with him (electron correlation
> >>> or not).
> >>> Wonder where he went? He seemed to have some serious
> >>> misconceptions. I may be wrong about this, but he'd
> >>> have to point out where we were wrong.
> Well I, for one, think he has a point, although he did not express
> it very well. Let me point out that I am not an ab initio
> quantum chemist myself, I'm a dynamicist who works with ab initio
> results so I hope TPH and MJR will set me straight when I get
> the details wrong as I probably will.
> >Our author had claimed that there was tomfoolery in chosing
> >special "basis sets" for doing the calculations on copper,
> >though he admitted that those gave answers that agreed with
> >experiment. At that point I noted that he seemed not to
> >understand what "basis sets" are and why the choice of
> >basis set only affected the efficiency of the computation
> >and not, if enough terms are taken, the result.
> But in practice "enough terms" are _never_ taken (except in
> something really simple like a two-electron system). The
> Hilbert space of a many-electron atom or molecule is
> very, very, very big. There is no hope of covering
> it with enough basis functions so that everything will
> converge.
Eric may have also been referring that the basis sets are optimized
with respect to experimental observables. Even so, there are
universal basis sets that are occasionally usable, so using
experimental data is not an absolute requirement. The optimized
basis sets give the most bang for the buck.
> >choice. In that same manner, quantum mechanicians choose
> >basis sets for their ability to mimic the known features
> >of the problem. It just makes the work easier.
> No, it makes the work _possible_. Quantum chemistry methods
> are chosen, at least in part, according to their track record
> for giving correct results in similar problems. There are
> heuristic arguments for these decisions: for example, if you
> want to treat a negative ion you should have diffuse functions
> in your atomic basis set because anion wave functions tend to
> spread out farther; if lone pairs are important to you, you
> should use a coupled-cluster method rather than MP perturbation
> theory because CC builds in pair correlations, etc. But these
> are heuristics, insights derived from physical intuition, chemical
> intuition, and the results of comparing theory with experiment
> on simpler problems. If you ignore the wisdom of the professionals,
> you don't get a calculation that converges slowly, you get a
> calculation that converges quite nicely to the wrong result.
> This is still more true in density functional theory: the
> exchange-correlation functionals that are in common use now seem
> to be based upon about 30% physical intuition, 40% accumulated
> experience with older functionals that didn't work, and 30% black magic.
Robert is referring to a type of method that mixes various methods
in a parameterized fashion.
Just wait until you see some of the 10 or 18 - parameter density functionals.
> So I do think it is appropriate to ask whether ab initio quantum
> chemistry calculations really derive their results from "first
> principles." Let me emphasize that I'm not singling out quantum
> chemistry in this respect; much the same is true of all cutting-
> edge research in theoretical physics and chemistry. Rigorous,
> a priori error estimates are rarely available for the most
> exciting experimental problems. If we restricted ourselves to
> problems where we could state error bounds, we'd probably be stuck
> somewhere around Beryllium.
> This is why questions of "prediction" in science are not simple.
> Ab initio quantum chemistry has reached the stage where disagreements
> between theory and experiment are not always resolved in favor of
> the experimentalist. (Yes, Tracy, I know you want to say something
> about methylene here. :-) ). That pragmatic criterion is what, for
> me, demonstrates the success of the field.
Sorry for the late relpy. I was forced to judge beer in Nashville
this weekend. And, no, I will not discuss the "m" word.
-----------------------------------------------------------------------
Tracy P. Hamilton |O Beer! O Hodgson, Guinness, Alsopp, Bass!
Building Manager, Alco Hall |Names that should be on every infants' tongue!
University of Ediacara | - Charles Stuart Calverly
-----------------------------------------------------------------------
-----------== Posted via Deja News, The Discussion Network ==----------
tpham...@my-dejanews.com
MJR <robe...@kemi.aau.dk> wrote:
> Robert Parson wrote:
> >
> > In article <70onua$7...@panix2.panix.com>, Paul J. Gans <ga...@panix.com>
wrote:
> >
> > >>> Hmm. All of this arose because of the article in the
> > >>> Scientific American. One of the authors was here posting
> > >>> some rather strange statements about quantum mechanics.
> > >>> Some of us didn't agree with him (electron correlation
> > >>> or not).
> > >>>
> > >>> Wonder where he went? He seemed to have some serious
> > >>> misconceptions. I may be wrong about this, but he'd
> > >>> have to point out where we were wrong.
> >
> > Well I, for one, think he has a point, although he did not express
> > it very well. Let me point out that I am not an ab initio
> > quantum chemist myself, I'm a dynamicist who works with ab initio
> > results so I hope TPH and MJR will set me straight when I get
> > the details wrong as I probably will.
>
> I have found the comments by Paul and Robert to be quite correct,
> gratifyingly so compared to the pointless drivel so often spewed out
> upon this newsgroup.
You wouldn't be referring to my posts, would you? :)
> >
> > >Our author had claimed that there was tomfoolery in chosing
> > >special "basis sets" for doing the calculations on copper,
> > >though he admitted that those gave answers that agreed with
> > >experiment. At that point I noted that he seemed not to
> > >understand what "basis sets" are and why the choice of
> > >basis set only affected the efficiency of the computation
> > >and not, if enough terms are taken, the result.
> >
> > But in practice "enough terms" are _never_ taken (except in
> > something really simple like a two-electron system). The
> > Hilbert space of a many-electron atom or molecule is
> > very, very, very big. There is no hope of covering
> > it with enough basis functions so that everything will
> > converge.
> >
>
> This point is so fundamental that it ought to be drilled into every
> first year student in computational science. Computers are stupid: once
> an input deck is submitted the machine will merrily follow the algorithm
> until the STOP signal is detected. The user is the one who is supposed
> to be smart.
That is, the STOP signal should be sent before the job is run.
> The problem is quite severe in chemistry, so much so that I have
> something of a standard speech I give every time an experimentalist
> comes to me after reading an article about the marvel advances in
> chemistry software; it runs along the lines of "what you are suggesting
> is *way* too difficult to accomplish in one lifetime, and the numbers I
> *can* compute are of no obvious relevence to your chemical problem. So,
> let's both think about this real hard and see if we can at least shine
> some new light on the problem and inspire a more clever line of
> investigation".
The software is too user friendly.
> Your comments on convergence might be misleading to the non expert;
> most people work with the integral/algebraic approximation to the
> Schroedinger equation rather than, strictly speaking, solving the
> differential equation form as this practically guarantees a numerically
> stable solution-which may of course be numerically precise garbage. The
> larger point is completely valid: as the basis is expanded the computed
> value converges towards the complete limit and the investigator has to
> think in terms of "what question am I trying to answer, does the method
> I use incorporate the essential physics, and just how much accuracy to I
> need?"
> As Paul observes, some 'problems' are in a sense completely fake.
> The whole idea of a 3d electron arises from the mean field approxima-
> tion; somewhat miraculously, this approach has been developed into a
> simple and powerful theory of chemical structure and bonding that
> researchers can manipulate within their heads to obtain important
> insights into the properties of matter. For this reason, not because
> electron shells are "real" in any formal sense, the standard theory of
> electronic structure has become one of the main pillars in the chemical
> edifice.
And then there are the complications of natural orbitals and
the extended Koopmans' theorem methods which can give ionization
potentials exactly.
The mere existence of Gaussian did more for theoreticians than
anything else. And Pople did a lot of development in
semi-empirical, quantum and density functional theory,
a trifecta that no one else has done.
Catty, catty! It wouldn't be a theoretician who thought that he
deserved it, would it? :)
> For a man whose contribution to the art is comparable to that of
> Walter Kohn in Density Functional Theory,
The selection of Walter Kohn was very interesting. A fundamental
advance to be sure, and also a "vote with the feet" as to where
quantum chemistry is headed.
> I would argue that Enrico
> Clemmenti or Per-Olav Lowdin should have been selected.
No. How about Clemens Roothaan?
> As a torpedo
> for the Scandnavian Mafia, I of course would have given the nod to Per-
> Olav; not only did he contribute a series of seminal papers, but by
> winkling funds from the US Air Force and training a raft of brilliant
> graduate students he did much to establish the modern school of
> electronic structure theory upon which all of our currect success is
> based.
> End of sermon.
Boo! Hisss!
Naked sabres at dawn, it is!
Actually, it points out how difficult it is to single out one
theoretician in chemistry.
Paul J. Gans
> End of sermon.
Well done! No argument from me here. And I'll not get involved
in the Clemmenti-Lowdin discussion.
One of the things that happens in net-discussions is that
they often either become too technical for other folks to
follow *or*, if simplified, run the risk of being criticized
for not being quite correct. I'll cheerfully plead guilty
to the latter. Which is why I made some dubious statements
that both MJR and RP were too polite to point out.
On the other hand, some things never change. Many years
ago, when Jules Moskowitz first came to NYU, complete with
one of the original gaussian codes, he pointed out to anyone
who would listen that most problems of interest were too
"hard" to do. It's 35 years later and most problems of
interest are still too hard to do... ;-)
In fairness though, the range of problems that *can* be done
is now fairly enormous, especially from the standpoint of
1963.
MJR
>
> On the other hand, some things never change. Many years
> ago, when Jules Moskowitz first came to NYU, complete with
> one of the original gaussian codes, he pointed out to anyone
> who would listen that most problems of interest were too
> "hard" to do. It's 35 years later and most problems of
> interest are still too hard to do... ;-)
>
The more things change and all that. I get a real kick out of
listening to stories of the olden days, it humanises the geniuses who
make possible our work today. Not unlike young initiates sitting around
the campfire with the tribal elders.
> In fairness though, the range of problems that *can* be done
> is now fairly enormous, especially from the standpoint of
> 1963.
>
> ------ Paul J. Gans [ga...@panix.com]
At least once a week I find myself thinking "we live in an age of
miracles".
**********************************************************************
To reply, replace the danish robersen with the norwegian roberson.
**********************************************************************
Robert Parson
> I seem to be particularly garrulous this morning, so at the risk of
>taking this deeper into the domain of soc.culture.comp-chem, I would
>like to ask what the smart guys think about Pople getting the Nobel Prize.
Granted, without the Gaussian codes no one outside an incestuous
>clique would have ever heard of electronic structure theory, but Pople
>has already been rewarded for that with his stock in Gaussian Inc. As a
>colleague, who shall for obvious reasons remain anonymous, remarked:
>Pople never had an original idea in his life.
And I know a few eminent dynamicists, and more than a few statistical
mechanics, who say the same about _all_ ab initio quantum chemists;
the stat mech guys in particular tend to rank quantum chemists slightly
below creationists on the intellectual food chain (though slightly above
their own competitors in their fields). You really don't want to start
playing this game; whole subfields have annihilated themselves in this
way. (Everybody nukes everybody else's proposal, until the funding
agencies simply scratch off the whole community.)
Many people forget that Nobel Prizes are not supposed to be given to
the smartest guy or the deepest thinker. They're given to the person
who makes a difference. By all accounts the guy who invented PCR is a
total butthead, but PCR changed the way molecular biologists do their
business. At least a dozen different groups could have discovered C-60,
but they didn't - Kroto and Smalley did, by sheer accident. In
this sense, designing Gaussian is like inventing PCR, or the
Scanning Tunnelling Microscope, or the Bubble Chamber, all of which
were recognized by Nobel awards.
Moreover, the Nobel committee strongly emphasizes _broad_ impact
across several fields. For a physical chemist to get a Nobel,
it's more important that he do something that influences organic,
inorganic and biochemists than to do a really sweet piece of physical
chemistry. That's why Rudy Marcus got the prize for his theory of
electron transfer reactions - everybody uses it, from analytical
electrochemists to computational biochemists.
Every time the Prize is announced, there are attempts to second-guess
the Committee. Many of my experimental colleagues think that
Dick Bernstein should have shared the 1987 prize, and I'm still
amazed that Mendeleev, G. N. Lewis, and Henry Eyring were never
recognized. Overall, though, they do a good job.
Now if only they'd get around to Dick Zare...
------
Robert
tpham...@my-dejanews.com
rpa...@spotNO.SPAMColorado.edu (Robert Parson) wrote:
> In article <3634F0...@kemi.aau.dk>, MJR <robe...@kemi.aau.dk> wrote:
> > I seem to be particularly garrulous this morning, so at the risk of
> >taking this deeper into the domain of soc.culture.comp-chem, I would
> >like to ask what the smart guys think about Pople getting the Nobel Prize.
> Granted, without the Gaussian codes no one outside an incestuous
> >clique would have ever heard of electronic structure theory, but Pople
> >has already been rewarded for that with his stock in Gaussian Inc. As a
> >colleague, who shall for obvious reasons remain anonymous, remarked:
> >Pople never had an original idea in his life.
> And I know a few eminent dynamicists, and more than a few statistical
> mechanics, who say the same about _all_ ab initio quantum chemists;
> the stat mech guys in particular tend to rank quantum chemists slightly
> below creationists on the intellectual food chain (though slightly above
> their own competitors in their fields). You really don't want to start
> playing this game; whole subfields have annihilated themselves in this
> way. (Everybody nukes everybody else's proposal, until the funding
> agencies simply scratch off the whole community.)
I wonder why they would say these things. A lot of ab initio quantum
chemistry is now well formulated so that even a person of my intelligence
can understand it, and maybe have a chance to contribute to its further
development. Is that a bad thing? I certainly prefer a paper or talk
that is well thought out and crystal clear to one that is osbscured
instead of illuminated by mathematics.
> Many people forget that Nobel Prizes are not supposed to be given to
> the smartest guy or the deepest thinker. They're given to the person
> who makes a difference. By all accounts the guy who invented PCR is a
> total butthead, but PCR changed the way molecular biologists do their
> business. At least a dozen different groups could have discovered C-60,
> but they didn't - Kroto and Smalley did, by sheer accident. In
> this sense, designing Gaussian is like inventing PCR, or the
> Scanning Tunnelling Microscope, or the Bubble Chamber, all of which
> were recognized by Nobel awards.
> Moreover, the Nobel committee strongly emphasizes _broad_ impact
> across several fields. For a physical chemist to get a Nobel,
> it's more important that he do something that influences organic,
> inorganic and biochemists than to do a really sweet piece of physical
> chemistry. That's why Rudy Marcus got the prize for his theory of
> electron transfer reactions - everybody uses it, from analytical
> electrochemists to computational biochemists.
> Every time the Prize is announced, there are attempts to second-guess
> the Committee. Many of my experimental colleagues think that
> Dick Bernstein should have shared the 1987 prize, and I'm still
> amazed that Mendeleev, G. N. Lewis, and Henry Eyring were never
> recognized. Overall, though, they do a good job.
Amazing that those guys were never chosen.
> Now if only they'd get around to Dick Zare...
-----------------------------------------------------------------------
Rex Nielson
-snip-
> Particles and waves are *concepts*. That is the whole point to
> "particle-wave" duality. Just because a particle can be modelled
> by a wave packet does not mean it *is* a wave packet. It is all a
> matter of which concept is more useful in which circumstance.
Maybe "particle-wave" duality is pointless, because the wave phenomena of
matter is more then a "concept" but an empirical, observable, macroscopic
fact. I've no doubt you've heard of the Bose-Einstein Condensate, in which
researchers have used lasers to cool matter to the point where its atoms
lose their individual identity and behave as a one "superatom" or a large
matterfield with directly observable wave properties. There's a raft of
articles about the subject at this URL:
<http://amo.phy.gasou.edu/bec.html/popular.html>
Doesn't this phenomena fly in the face of the Heisenburg's uncertainty
principle. Knowledge of the matterfield's individual "parts" is not
subject to meaningful probablistic interpretation becuase the matter-wave
is directly observable.
Parson has a colorado.edu addresss, where evidently some of this research
is taking place. Maybe he should walk down the hall and tell them they'd
better stop or else they'll disprove QM.
> > >>I'm supremely unqualified to comment on the plausibility of theories of
> > >>physics by interpretation of experiment or in terms mathematical proofs,
> > >>but I think I can see a curious pattern at work. A few biologists today
> > >>ridicule neoDarwinism for its invitation to describe evolution as a story
> > >>about the adventures of the genes. I think the inspiration for this comes
> > >>ultimately from mechanistic and atomistic assumptions that run deep in
> > >>science and are not held by biologists alone. Perhaps future physicists
> > >>will make fun of their predecessors for having described material
> >>phenomena
> > >>as 'the adventures of the particles'.
>
> I don't see the pattern you are referring to. All I see are
> refinements of or additions to still good theories.
QM and neoDarwinism attribute the ultimate reality of the physical world
and of life to the goings on of subatomic and molecular particles.
NeoDarwinism has led to some interesting research, but it relies dubious
stories of random molecular mutation and phenotypic adaptation. These
conjectures are justified on the basis of our historical
ignorance---evolution would have taken place in an unobservable realm. QM
seems to follow the same pattern, in that it is based on conjectures about
particle phenomena that according to its principles occur in an
unobservable realm. The biologist Brian Goodwin argues that the phenomena
responsible for evolution isn't chance historical accumulation, forever
obscured by the past, but rational generative principles. Perhaps the
phenomena responsible for the material world is not necessarily obscured by
its smallness. Perhaps these sciences based on ignorance can have their
useful parts removed and be consigned to the trash heap.
>
> > Well okay, thorium to. But you'll have to concede that the heating of the
> > earth and subsequent tectonic activity is the result of the radioactive
> > decay of the heavy elements, unless you want to follow Lord Kelvin and say
> > the earth considerably younger then is thought today.
>
> But is tectonic activity necessary for life?
It's impossible to say for sure what life could and couldn't be.
Speculation about the possibility of *simple* forms of life on geologically
dead worlds is common. It seems certain however that because of the
tectonic cycle's part in generating and maintaining our atmosphere with its
integrated role with the hydrologic cycle, that without it there wouldn't
be an environment suitable for life-as-we-know-it.
>
> > As for the noble gases, Denton suggests they are necessary by-products of
> > the atom-building rules. Perhaps seeming aberrations can be justified in
> > the context of a larger scheme. One can, of course, justify anything with
> > this kind of reasoning and with recourse to the explanation of "higher law"
> > there is potential for abuse (e.g. God made neon for lightbulbs). But if
> > evidence is clear that the universe may be governed by higher laws,
> > students of nature are obliged to pursue and discover them, even if this
> > contradicts strongly held conventions.
>
> Until we could be absoulutely certain about what impact a slight change
> in some fundamental aspect of the universe (Or if they are even
> variable, or even other universes possible) then "how nicely things seem
> to fit together" only supports various idle speculations.
There are plenty of cosmologists who feel "how nicely things seem to fit
together" are worthy of their speculations. Linde thinks our universe is a
mere bubble in a giant foamy bubble bath of an ultimate reality.
Presumably the other bubbles have qualities less convenient for life.
Smolin postulates the severely illogical idea the universes actually birth
each other via black holes and that our universe, and not just us, is a
product of natural selection. This type of speculation is at a distinct
disadvantage to those who would support design because it relies on
conjectures about what lies outside our universe where observable facts
would be (for obvious reasons) considerable harder to come by then for
those who look for meaning and pattern within our own universe.
-RN
wf...@enter.netxx
wrote:
>
>QM and neoDarwinism attribute the ultimate reality of the physical world
>and of life to the goings on of subatomic and molecular particles.
huh? izzat right?? dont recall QM being mentioned in 'origin of
species'
>NeoDarwinism has led to some interesting research, but it relies dubious
>stories of random molecular mutation and phenotypic adaptation
what stories?? methinks you have science confused with the fairy tales
of designers creating fully formed species ex nihilo
.. These
>conjectures are justified on the basis of our historical
>ignorance---evolution would have taken place in an unobservable realm.
except, of course, we see evolution taking place today. creationists
dont like observation or evidence as part of science because it
contradicts their view of the world, but thems the facts.
QM
>seems to follow the same pattern, in that it is based on conjectures about
>particle phenomena that according to its principles occur in an
>unobservable realm.
really? whats unobservable about QM??
.. Perhaps these sciences based on ignorance can have their
>useful parts removed and be consigned to the trash heap.
what is the 'useful part' you refer to? oh...i forgot...all the parts
that agree with your religion...sorry
yes, creationists do practice cafeteria science...
>
>Smolin postulates the severely illogical idea the universes actually birth
>each other via black holes and that our universe, and not just us, is a
>product of natural selection.
why is it illogical? heilman doesnt say. other than his religious bias
he offers no explanation...
This type of speculation is at a distinct
>disadvantage to those who would support design because it relies on
>conjectures about what lies outside our universe where observable facts
>would be (for obvious reasons) considerable harder to come by then for
>those who look for meaning and pattern within our own universe.
except smolin bases his speculation on scientific principles.
creationism is based on magic....as is design.
Robert Parson
<tpham...@my-dejanews.com> wrote:
>In article <715ip8$2...@peabody.colorado.edu>,
> rpa...@spotNO.SPAMColorado.edu (Robert Parson) wrote:
>> In article <3634F0...@kemi.aau.dk>, MJR <robe...@kemi.aau.dk> wrote:
>
>> >Pople never had an original idea in his life.
>
>> mechanics, who say the same about _all_ ab initio quantum chemists;
>> the stat mech guys in particular tend to rank quantum chemists slightly
>> below creationists on the intellectual food chain
>I wonder why they would say these things.
1. Professional jealousy on the part of the dynamicists: quantum
chemistry "matured" earlier than dynamics, so that quantum chemists
were saying useful things about experimentally significant problems when
dynamicists were still playing with the collinear H+H2 reaction.
Let me note that the same "eminent dynamicists" of whom I was thinking
are (or were) completely clueless about what electronic structure
actually involves - as I realized once I was forced to learn some
ab initio theory myself.
2. The arrogant attitudes of a _few_ quantum chemists, who acted as if
quantum chemistry was coextensive with theoretical chemistry - that
once they'd gotten the potential surface, all the rest was trivial.
Such people shall, of course, remain nameless except for the ones
from southern California whose names begin with G and end with D. :-)
> A lot of ab initio quantum
>chemistry is now well formulated so that even a person of my intelligence
>can understand it, and maybe have a chance to contribute to its further
>development. Is that a bad thing? I certainly prefer a paper or talk
>that is well thought out and crystal clear to one that is osbscured
>instead of illuminated by mathematics.
3. I agree completely - the problem is that this was NOT the case
15 years ago. Back then, if you went to an ab initio talk you were
likely to be confronted with strings of undefined acronyms and slide
after slide consisting of tables of numbers, with little or not
motivation given for what really going on.
That's why Szabo and Ostlund was such a revelation for many of us who
weren't brought up in the ab initio tradition. They not only defined
all those ghastly acronyms, they let us in to see just what was going
on by illustrating how it all worked in minimal basis. No one had
ever done that before! (I hope that dynamicists Schatz and Ratner have repaid
the debt in their own wonderful book, _Quantum Mechanics in Chemistry_.)
Nowadays, the schism between dynamicists and ab initio quantum chemists
is pretty well gone, mostly because (starting with Car-Parrinello 1985)
the boundary between the subdisciplines has dissolved away. The interesting
dynamics problems now, in my not-at-all-humble-opinion, are those in which
electronic and nuclear motion get hopelessly entangled. But you know
that if you've been to our web page :-).
As for the statistical mechanics, though, they're still hopeless. :-)
Robert Parson
Rex Nielson <rnie...@xmission.com> wrote:
>In article <712gek$u69$1...@nnrp1.dejanews.com>, tpham...@my-dejanews.com wrote:
>
>-snip-
>> Particles and waves are *concepts*. That is the whole point to
>> "particle-wave" duality. Just because a particle can be modelled
>> by a wave packet does not mean it *is* a wave packet. It is all a
>> matter of which concept is more useful in which circumstance.
>
>
>Maybe "particle-wave" duality is pointless, because the wave phenomena of
>matter is more then a "concept" but an empirical, observable, macroscopic
>fact. I've no doubt you've heard of the Bose-Einstein Condensate, in which
>researchers have used lasers to cool matter to the point where its atoms
>lose their individual identity and behave as a one "superatom" or a large
>matterfield with directly observable wave properties.
As Nathan Urban says, this does not add anything to the basic
wave-particle duality demonstrated in experiments on single particles
(e.g. electron diffraction). It merely demonstrates (beautifully) the
applicability of the same concepts to collections of thousands or
millions of atoms, all doing their wave-particle duality stuff in synch.
By the way, Einstein's original derivation of BEC made no use of
wave-particle duality (because it hadn't been discovered yet). It's
a purely statistical argument, of the sort which Einstein could
handle better than anyone else. Nowadays the "phase space density"
factor appearing in Einstein's formula is interpreted in terms of
overlapping deBroglie waves (and this is a very enlightening
interpretation), but Einstein didn't need this interpretation to
get to the result!
>Doesn't this phenomena fly in the face of the Heisenburg's uncertainty
>principle.
Absolutely not! In fact, the original BEC experiment provides a
beautiful illustration of the uncertainty principle. The condensate
images were produced by shutting off the trap and letting the gas
expand. They thus give you a picture of the velocity distribution
in the condensate - what quantum mechanics calls the "momentum space
wave function". The images are elliptical rather than spherical,
because the trap is anisotropic. If the gas were classical, this
wouldn't matter, the momentum distribution would be isotropic
(the kinetic energy distribution of a classical gas in an anisotropic
well is isotropic, according to Boltzmann; however, the kinetic energy
distribution of a quantum gas is _not_, because squeezing down one
of the spatial dimensions makes the corresponding momentum direction
become more uncertain.)
>
>Parson has a colorado.edu addresss, where evidently some of this research
>is taking place. Maybe he should walk down the hall and tell them they'd
>better stop or else they'll disprove QM.
Actually it's down three flights of stairs. :-)
BTW, I don't work on BEC, or anything close, but I've probably heard
more BEC talks than the rest of this newsgroup added together. It's
a beautiful experiment, but fully within the domain of orthodox
quantum mechanics.
------
MJR
>
> In article <717c6v$aih$1...@nnrp1.dejanews.com>,
> <tpham...@my-dejanews.com> wrote:
> >In article <715ip8$2...@peabody.colorado.edu>,
> > rpa...@spotNO.SPAMColorado.edu (Robert Parson) wrote:
> >> In article <3634F0...@kemi.aau.dk>, MJR <robe...@kemi.aau.dk> wrote:
> >
> >> >Pople never had an original idea in his life.
> >
> >> mechanics, who say the same about _all_ ab initio quantum chemists;
> >> the stat mech guys in particular tend to rank quantum chemists slightly
> >> below creationists on the intellectual food chain
Not without some justification, I eagerly concede. As much as I
admire the technical sophistication of Gaussian 94, the ease of input
lures many people in over their heads. To my mind their is a clear
distinction between the applied mathematicians who develope new methods,
the ab initio jocks who explore the practical behaviour of the
technology, and the mass of chemists doing industrial scale production
runs. One can draw an anology to NMR, with the distinction between
those doing high magnetic field research and the mass of organic
chemists using it as an analytical tool. Great work is being done on
all fronts, but whereas a bogus proton spectrum is obvious to any
technician a bogus calculation does not necessarily send up warning
flags.
Pardon my leaping back in again. My server seems to have gone
screwy, hiding from me a number of interesting posts.
> >I wonder why they would say these things.
>
> 1. Professional jealousy on the part of the dynamicists: quantum
> chemistry "matured" earlier than dynamics, so that quantum chemists
> were saying useful things about experimentally significant problems when
> dynamicists were still playing with the collinear H+H2 reaction.
>
> Let me note that the same "eminent dynamicists" of whom I was thinking
> are (or were) completely clueless about what electronic structure
> actually involves - as I realized once I was forced to learn some
> ab initio theory myself.
Don't forget that academics are a notoriously hissy lot.
>
> 2. The arrogant attitudes of a _few_ quantum chemists, who acted as if
> quantum chemistry was coextensive with theoretical chemistry - that
> once they'd gotten the potential surface, all the rest was trivial.
> Such people shall, of course, remain nameless except for the ones
> from southern California whose names begin with G and end with D. :-)
>
Har Har Har. This wouldn't be the modest genius who named a
correlated valence bond method after himself, would it?
> > A lot of ab initio quantum
> >chemistry is now well formulated so that even a person of my intelligence
> >can understand it, and maybe have a chance to contribute to its further
> >development. Is that a bad thing? I certainly prefer a paper or talk
> >that is well thought out and crystal clear to one that is osbscured
> >instead of illuminated by mathematics.
>
> 3. I agree completely - the problem is that this was NOT the case
> 15 years ago. Back then, if you went to an ab initio talk you were
> likely to be confronted with strings of undefined acronyms and slide
> after slide consisting of tables of numbers, with little or not
> motivation given for what really going on.
Ditto. I have tatooed on my forehead the old dictum "the Purpose
of Computing is Insight not Numbers". To put this in perspective,
though, the late eighties was the time when algorithms were being
implemented for things like gradients and hessians; until that was
finished, getting a PES for anything more complicated that H2+H was a
nightmare.
>
> That's why Szabo and Ostlund was such a revelation for many of us who
> weren't brought up in the ab initio tradition. They not only defined
> all those ghastly acronyms, they let us in to see just what was going
> on by illustrating how it all worked in minimal basis. No one had
> ever done that before! (I hope that dynamicists Schatz and Ratner have repaid
> the debt in their own wonderful book, _Quantum Mechanics in Chemistry_.)
>
> Nowadays, the schism between dynamicists and ab initio quantum chemists
> is pretty well gone, mostly because (starting with Car-Parrinello 1985)
> the boundary between the subdisciplines has dissolved away. The interesting
> dynamics problems now, in my not-at-all-humble-opinion, are those in which
> electronic and nuclear motion get hopelessly entangled. But you know
> that if you've been to our web page :-).
The most straightforward way to treat this problem is to chose your
system wisely, so that all those troublesome none-BO terms can be swept
under the rug :-O
Rex Nielson
> In article <rnielson-ya0240800...@news.xmission.com>,
rnie...@xmission.com (Rex Nielson) wrote:
>
> > Maybe "particle-wave" duality is pointless, because the wave phenomena of
> > matter is more then a "concept" but an empirical, observable, macroscopic
> > fact.
>
> So is the particle phenomenon of matter.
>
> > [BECs, macroscopically observable matter wave interference]
>
> > Doesn't this phenomena fly in the face of the Heisenburg's uncertainty
> > principle.
>
> No.
>
> Furthermore, I don't see what the relevance of BECs to wave-particle
> duality is. Are you trying to argue that macroscopically observable
> wavelike properties means that matter admits a wave description but not
> a particle one? Remember, matter interference has been observed for a
> long time, e.g., you can do an electron two-slit experiment just like you
> can do a light two-slit experiment. Sure, you get interference effects,
> but at the lowest level you're observing discrete quanta.
The wave properties of BECs are macroscopic and would presumably be
objective, exhibiting the same characteristics under a variety of
experimental perspectives. As I understand it, subjectivity is integral to
QM's description of the wave properties of matter. According to the
theory, matter can appear to be a wave or it can appear to a particle
corresponding with the experimental conditions. Not so, when matter is
cold enough, then it becomes a "superatom" something that looks not like a
pool-ball but a matterfield whose wave properties can evidently be
observed with the naked eye [Sci. Amer. Mar '98 p.40]. How can matter be a
wave and a particle at the same time... The answer QM gave was that 'pure'
matter was shrouded by its smallness and the very act of looking at it
would affect its appearance, so it was as far as we were concerned it could
be both wave and particle. QM's probabilistic *subjective* interpretation
of matter is legitimated by this indeterminacy. But is this a valid
inquiry into the true nature of matter? With *objective* wave phenomena in
matter now being observed it would seem necessary to call QM's presumptions
into question. Perhaps QM is not a description of reality only a handy
mathematical approximation.
>
> > Knowledge of the matterfield's individual "parts" is not
> > subject to meaningful probablistic interpretation becuase the matter-wave
> > is directly observable.
>
> What does that have to do with the HUP?
>
> > Smolin postulates the severely illogical idea the universes actually birth
> > each other via black holes
>
> This is rather speculative, but far from illogical. In fact, it provides
> a nice solution to the black hole information loss paradox. You seem
> to have some fundamental misunderstanding of the word "illogical".
> It's only illogical if it is internally inconsistent.
Smolin's logic is at least severely circular. According to natural
selection various qualities are supposedly chosen through interaction with
the environment, since there could be no environment for a universe (unless
you want to begin the circle now by saying it would exists in a larger
universe) the various qualities of the spawned universes would have to come
from their own source. This source would be black holes or singularity or
whatever. Apparently the only reason you can give by this theory for a
particular quality of our universe is that it exists for and by a black
hole. Let me try one. Why does our universe have suns? Because suns can
collapse into black holes and black holes spawn more universes giving a
statistical predominance in the grand reality to those universes with suns,
so our sun is to be expected. That's the kind of tautology that would make
a Sociobiologist blush. With black holes that somehow produce universes
like the ones in which they were formed, it would be just as easy to
imagine abortive universes that soon collapse into a handful of
singularities flooding the metaverse (how's that word?) by breeding
quickly, whereby our old universe would become very unexpected again.
Smolin's theory seems weird to me.
-RN
>
> > and that our universe, and not just us, is a
> > product of natural selection. This type of speculation is at a distinct
> > disadvantage to those who would support design because it relies on
> > conjectures about what lies outside our universe where observable facts
> > would be (for obvious reasons) considerable harder to come by then for
> > those who look for meaning and pattern within our own universe.
>
> It nonetheless has consequences that are testable within this universe.
wf...@enter.netxx
wrote:
>
>Smolin's logic is at least severely circular. According to natural
>selection various qualities are supposedly chosen through interaction with
>the environment, since there could be no environment for a universe
ah, the literalist pushes an analogy to the extreme.
AFAIK smolin never proposed that 'natural selection' was EXACTLY like
darwin's selection. its an analogy, boy! it aint exact. it aint a
perfect one, and it seems that those who have a literal frame of mind
(gee what a surprise for a creationist) are having conniptions with
the whole idea.
Paul J. Gans
>In article <717c6v$aih$1...@nnrp1.dejanews.com>,
> <tpham...@my-dejanews.com> wrote:
>>In article <715ip8$2...@peabody.colorado.edu>,
>> rpa...@spotNO.SPAMColorado.edu (Robert Parson) wrote:
>>> In article <3634F0...@kemi.aau.dk>, MJR <robe...@kemi.aau.dk> wrote:
>>
>>> >Pople never had an original idea in his life.
>>
>>> mechanics, who say the same about _all_ ab initio quantum chemists;
>>> the stat mech guys in particular tend to rank quantum chemists slightly
>>> below creationists on the intellectual food chain
>>I wonder why they would say these things.
> 1. Professional jealousy on the part of the dynamicists: quantum
> chemistry "matured" earlier than dynamics, so that quantum chemists
> were saying useful things about experimentally significant problems when
> dynamicists were still playing with the collinear H+H2 reaction.
> Let me note that the same "eminent dynamicists" of whom I was thinking
> are (or were) completely clueless about what electronic structure
> actually involves - as I realized once I was forced to learn some
> ab initio theory myself.
> 2. The arrogant attitudes of a _few_ quantum chemists, who acted as if
> quantum chemistry was coextensive with theoretical chemistry - that
> once they'd gotten the potential surface, all the rest was trivial.
> Such people shall, of course, remain nameless except for the ones
> from southern California whose names begin with G and end with D. :-)
>> A lot of ab initio quantum
>>chemistry is now well formulated so that even a person of my intelligence
>>can understand it, and maybe have a chance to contribute to its further
>>development. Is that a bad thing? I certainly prefer a paper or talk
>>that is well thought out and crystal clear to one that is osbscured
>>instead of illuminated by mathematics.
> 3. I agree completely - the problem is that this was NOT the case
> 15 years ago. Back then, if you went to an ab initio talk you were
> likely to be confronted with strings of undefined acronyms and slide
> after slide consisting of tables of numbers, with little or not
> motivation given for what really going on.
> That's why Szabo and Ostlund was such a revelation for many of us who
> weren't brought up in the ab initio tradition. They not only defined
> all those ghastly acronyms, they let us in to see just what was going
> on by illustrating how it all worked in minimal basis. No one had
> ever done that before! (I hope that dynamicists Schatz and Ratner have repaid
> the debt in their own wonderful book, _Quantum Mechanics in Chemistry_.)
> Nowadays, the schism between dynamicists and ab initio quantum chemists
> is pretty well gone, mostly because (starting with Car-Parrinello 1985)
> the boundary between the subdisciplines has dissolved away. The interesting
> dynamics problems now, in my not-at-all-humble-opinion, are those in which
> electronic and nuclear motion get hopelessly entangled. But you know
> that if you've been to our web page :-).
> As for the statistical mechanics, though, they're still hopeless. :-)
Hey!!!!! We S&M types matured even earlier than any of the
quantum types. Of course, some might say that the field
ossified before quantum mechanics reached its current state,
but I wouldn't go *that* far.
I've never really been a quantum mechanic. Ratner tried to
convert me back when he was at NYU, but I was headstrong. ;-)
Paul J. Gans
[...]
> The most straightforward way to treat this problem is to chose your
>system wisely, so that all those troublesome none-BO terms can be swept
>under the rug :-O
>>
>> As for the statistical mechanics, though, they're still hopeless. :-)
Hey, I *know* what BO stands for.... Does that count? ;-)
tpham...@my-dejanews.com
> rnie...@xmission.com (Rex Nielson) wrote:
> >
> > > matter is more then a "concept" but an empirical, observable, macroscopic
> > > fact.
> >
> >
> > > [BECs, macroscopically observable matter wave interference]
> >
> > > principle.
> >
> >
> > Furthermore, I don't see what the relevance of BECs to wave-particle
> > duality is. Are you trying to argue that macroscopically observable
> > wavelike properties means that matter admits a wave description but not
> > a particle one? Remember, matter interference has been observed for a
> > long time, e.g., you can do an electron two-slit experiment just like you
> > can do a light two-slit experiment. Sure, you get interference effects,
> > but at the lowest level you're observing discrete quanta.
> The wave properties of BECs are macroscopic and would presumably be
> objective, exhibiting the same characteristics under a variety of
> experimental perspectives. As I understand it, subjectivity is integral to
> QM's description of the wave properties of matter. According to the
> theory, matter can appear to be a wave or it can appear to a particle
> corresponding with the experimental conditions. Not so, when matter is
> cold enough, then it becomes a "superatom" something that looks not like a
> pool-ball but a matterfield whose wave properties can evidently be
> observed with the naked eye [Sci. Amer. Mar '98 p.40].
If I may interrupt, the "wave" in the wave nature is a *probability*
amplitude. This is not the same as a wave in water or other medium.
We say there is wave behavior *if there is interference*. The wave
function must also satisfy certain symmetry requirements, depending
on whether they are bosons or fermions.
The Bose-Einstein condensate is a collection of atoms that behave
on the whole as bosons, and below a certain temperature all of the
atoms go into the ground state, in a phase transition. This is not
an interference effect (I could be wrong!) but is still a quantum
effect on a macroscopic scale.
This is much like exchange phenomena in fermions being really just a
manifestation of antisymmetry, not a physical effect, but with
physical consequences.
> How can matter be a
> wave and a particle at the same time...
That is why I prefer to think of atoms as particles, and the wave
nature as just "quantum behavior".
> The answer QM gave was that 'pure'
> matter was shrouded by its smallness and the very act of looking at it
> would affect its appearance,
It is very important to note that this is true *even if the act
of looking at the particle does not physically disturb the system.*
I would say it affects its wave function rather than its appearance.
> so it was as far as we were concerned it could
> be both wave and particle.
The concept is not necessary, and I maintain it makes no sense.
> QM's probabilistic *subjective* interpretation
> of matter is legitimated by this indeterminacy. But is this a valid
> inquiry into the true nature of matter? With *objective* wave phenomena in
> matter now being observed it would seem necessary to call QM's presumptions
> into question. Perhaps QM is not a description of reality only a handy
> mathematical approximation.
Parts of this are incoherent. I find it odd that the BEC is a prediction from
quantum phenomena, yet you say that the BEC is a reason to call quantum
theory into question. The issues of interpretation are quite separate
from whether QM is a description of reality (from the computational
point of view).
[snip remainder]
Perhaps you could be a little clearer on your point. Is it
that matter is not a particle, but just a wave? Just because it
has quantum behavior is not sufficient, because the waves are
probability functions.
-----------------------------------------------------------------------
Tracy P. Hamilton |Have you ever noticed? Anybody going slower
Building Manager, Alco Hall |than you is an idiot, and anyone going
University of Ediacara |faster than you is a maniac. * George Carlin
-----------------------------------------------------------------------
Robert Parson
>Hey!!!!! We S&M types matured even earlier than any of the
>quantum types. Of course, some might say that the field
>ossified before quantum mechanics reached its current state,
>but I wouldn't go *that* far.
What I was thinking of, was that Stat Mech was the last holdout of
the pure paper-and-pencil theorists, and many of those guys believed
that computational theorists were an intellectually inferior breed.
(Apparently they'd never studied Gauss, Fermi or Chandrasekhar, who'd
been doing "computational physics" back before there _were_ computers.)
I still detected some of this attitude when I was a grad student
in the 1980's; it must have been much worse in the early 1960's.
(For the benefit of others reading this thread, Paul Gans was one of
the first SM theorists to embrace computation.)
------
Robert
tpham...@my-dejanews.com
Rex Nielson
> > The wave properties of BECs are macroscopic and would presumably be
> > objective,
>
> Macroscopicity has NOTHING to do with "objectiveness" or "subjectiveness".
> Quantum mechanics and the interpretation of quantum mechanics are
> not qualitatively different on a macroscopic scale than they are on a
> microscopic scale.
>
> > exhibiting the same characteristics under a variety of
> > experimental perspectives. As I understand it, subjectivity is integral to
> > QM's description of the wave properties of matter.
>
> The formulation of quantum mechanics makes no use of the term
> "subjectivity". Whether or not "subjectivity" has anything to do with
> quantum mechanics depends entirely on what you mean by the word, and I
> seriously doubt you can place that term on a rigorous physical grounding.
>
> > According to the
> > theory, matter can appear to be a wave or it can appear to a particle
> > corresponding with the experimental conditions.
>
> In a vague sense.
>
> > Not so, when matter is
> > cold enough, then it becomes a "superatom" something that looks not like a
> > pool-ball but a matterfield whose wave properties can evidently be
> > observed with the naked eye
>
> As I said before, a Bose-Einstein condensate is no more and no less
> wavelike OR particle-like than any other quantum phenomenon, e.g.,
> diffraction of light or electrons through a slit.
>
> > How can matter be a
> > wave and a particle at the same time... The answer QM gave was that 'pure'
> > matter was shrouded by its smallness
>
> "Smallness" has nothing to do with it. YOU are just as much a wave
> and just as much a particle as a Bose-Einstein condensate. It's just
> that you can't distinguish between individual particle states in a BEC
> without destroying the condensate.
>
> > QM's probabilistic *subjective* interpretation
> > of matter is legitimated by this indeterminacy. But is this a valid
> > inquiry into the true nature of matter? With *objective* wave phenomena
>
> BECs are no more and no less "objective" than any other quantum
> phenomenon.
>
> > > > Smolin postulates the severely illogical idea the universes
actually birth
> > > > each other via black holes
>
> > > This is rather speculative, but far from illogical. In fact, it provides
> > > a nice solution to the black hole information loss paradox. You seem
> > > to have some fundamental misunderstanding of the word "illogical".
> > > It's only illogical if it is internally inconsistent.
>
> > Smolin's logic is at least severely circular.
>
> Only if you severely misunderstand it.
>
> > According to natural
> > selection various qualities are supposedly chosen through interaction with
> > the environment,
>
> Smolin's "cosmological natural selection" is not natural selection in
> this sense, as I have pointed out before.
>
> > since there could be no environment for a universe (unless
> > you want to begin the circle now by saying it would exists in a larger
> > universe) the various qualities of the spawned universes would have to come
> > from their own source. This source would be black holes or singularity or
> > whatever.
>
> So?
>
> > Apparently the only reason you can give by this theory for a
> > particular quality of our universe is that it exists for and by a black
> > hole.
>
> The point of the theory is to determine which _physical constants_
> are _likely_. It isn't meant to describe every quality of our universe.
>
> > Let me try one. Why does our universe have suns? Because suns can
> > collapse into black holes and black holes spawn more universes giving a
> > statistical predominance in the grand reality to those universes with suns,
> > so our sun is to be expected. That's the kind of tautology that would make
> > a Sociobiologist blush.
>
> That's not a "tautology", that's a PREDICTION. The theory _predicts_
> stars. We observe them. That is a confirmation of the theory.
>
> This is not as circular as you think it is: you don't have to have a
> universe with stars in order to produce a universe with stars.
>
> > With black holes that somehow produce universes
> > like the ones in which they were formed, it would be just as easy to
> > imagine abortive universes that soon collapse into a handful of
> > singularities flooding the metaverse (how's that word?) by breeding
> > quickly, whereby our old universe would become very unexpected again.
>
> Those universes are possible, but not particularly relevant.
> The "population" of universes can have multiple "species"; each
> species is located at a local maximum of black hole production in the
> parameter space. All Smolin's theory says is that we are likely to
> find ourselves at a maximum, not necessarily the _absolute_ maximum.
> (i.e., that we should inhabit _a_ likely universe, not necessarily _the
> most_ likely universe.) If you have a peak in the parameter space, then
> there will always be a stable, self-sustaining population located there,
> no matter what is going on elsewhere.
>
> Personally, I think you need a touch of the anthropic principle here --
> you can rule out some of the peaks by virtue of them not being able
> to produce us, and then consider the predictions of parameters for
> the remaining peaks. That still counts as a prediction of a physical
> theory -- nobody said that theories had to make _unique_ predictions.
> That's simply the way things go in a statistical theory.
>
> > Smolin's theory seems weird to me.
>
> I'll grant you the weird part, but that's a far cry from being "severely
> illogical". Maybe you should read his book. But better yet, why don't
> you start reading some physics books instead of philosophy books.
> Philosophers who fancy themselves knowledgeable about the universe
> have a way of falling on their faces when they try to discuss science.
> You seem very confused about quantum mechanics. I'd suggest that you
> entirely drop the issue of "subjectivity" vs. "objectivity"; it's not
> particularly relevant to the predictions of the theory and certainly
> doesn't distinguish between "microscopic" and "macroscopic" or "particle"
> and "wave".
I may as well come clean, I have read a book about physics. In fact I've
been drawing heavily from theoretical physicist Mendel Sachs' book for
non-experts _Einstein Versus Bohr_ . In which he makes a point of
emphasizing the distinction between subjectivity and objectivity. He says
"irreducible subjectivity in the role of measuring apparatus as a
fundamental ingredient in our understanding of matter"---in terms of
Quantum mechanics that is. He negatively contrasts this with the
fundamentally "objective" theory of relativity. He says for this and other
reasons QM and the theory of relativity are incompatible. This would seem
to put modern physics in quite a pickle, don't you think? He summarizes
his views at this web sight:
<http://electron.physics.buffalo.edu/professors/sachs.html>
-RN
"I want to know how God created the world. I am not interested in this or
that phenomena, in the spectrum of this or that element; I want to know his
thoughts; the rest are details" -Albert Einstein
Rex Nielson
> If I may interrupt, the "wave" in the wave nature is a *probability*
> amplitude. This is not the same as a wave in water or other medium.
> We say there is wave behavior *if there is interference*. The wave
> function must also satisfy certain symmetry requirements, depending
> on whether they are bosons or fermions.
>
> The Bose-Einstein condensate is a collection of atoms that behave
> on the whole as bosons, and below a certain temperature all of the
> atoms go into the ground state, in a phase transition. This is not
> an interference effect (I could be wrong!) but is still a quantum
> effect on a macroscopic scale.
>
> This is much like exchange phenomena in fermions being really just a
> manifestation of antisymmetry, not a physical effect, but with
> physical consequences.
Researches have now gotten interference effects from the Bose-Einstein
condensate. This is from a recent article from Science magazine online:
[quote]
First, he and his colleagues created two condensates by beaming a laser up
through the middle of their magnetic trap.
The laser light repelled the atoms and split the condensate into two
distinct halves. For this test, there was no need to pulse the condensates
out of the trap; instead, the group just turned off the trap and let them
"free fall"--expand into the surrounding vacuum. The condensates swelled
until they overlapped and interfered, demonstrating the atomic version of
the bright and dark fringes in an interference pattern.
"The density of the overlapping region is modulated," says Ketterle. "Every
15 microns, we have matter, no matter, matter, no matter. Now we just shine
some light onto the pattern and see this shadow with black-and-white
stripes." Says Burnett, "It's not just a little crappy demonstration but a
big, juicy interference pattern."
[end quote]
The URL for this article is:
<http://www.sciencemag.org/feature/data/bose.shl>
These researchers have exploited the BEC's wave properties to make a
primitive "atom laser".
>
> > How can matter be a
> > wave and a particle at the same time...
>
> That is why I prefer to think of atoms as particles, and the wave
> nature as just "quantum behavior".
Isn't "quantum behavior" often seen as pretty topsy-turvy? At the quantum
level things don't seem to necessarily follow cause and effect
relations---electrons unpredictably jumping to lower energy states, photons
arbitrarily being created and destroyed. Objects don't suddenly change, or
blink in and out of existence in the larger world we know and live in.
Wouldn't the appearance of material wave phenomena on the macroscopic scale
we are familiar with call into question the assumption that the wave nature
of matter can be attributed to "quantum behavior"?
>
> > The answer QM gave was that 'pure'
> > matter was shrouded by its smallness and the very act of looking at it
> > would affect its appearance,
>
> It is very important to note that this is true *even if the act
> of looking at the particle does not physically disturb the system.*
> I would say it affects its wave function rather than its appearance.
>
> > so it was as far as we were concerned it could
> > be both wave and particle.
>
> The concept is not necessary, and I maintain it makes no sense.
I was attempting to say that according to QM, matter is attributed to
particles, even though particles can't be directly seen. The reason
particles can't be directly observed is attributed fundamental limitations
on our ability to measure. This would necessitate a probabilistic
interpretation.
>
> > QM's probabilistic *subjective* interpretation
> > of matter is legitimated by this indeterminacy. But is this a valid
> > inquiry into the true nature of matter? With *objective* wave phenomena in
> > matter now being observed it would seem necessary to call QM's presumptions
> > into question. Perhaps QM is not a description of reality only a handy
> > mathematical approximation.
>
> Parts of this are incoherent. I find it odd that the BEC is a prediction from
> quantum phenomena, yet you say that the BEC is a reason to call quantum
> theory into question. The issues of interpretation are quite separate
> from whether QM is a description of reality (from the computational
> point of view).
>
> [snip remainder]
>
> Perhaps you could be a little clearer on your point. Is it
> that matter is not a particle, but just a wave? Just because it
> has quantum behavior is not sufficient, because the waves are
> probability functions.
All the articles I've read about the BEC describe the it's unique
attributes as QM effects writ large. Wave behavior in matter is
automatically associated with Quantum Mechanics, but this seems to be
merely convention at work because probabilistic activity of particles isn't
the only way interpret wave behavior in matter. According to what I've
read conventional QM wasn't even the original way to interpret wave
behavior in matter:
[quote]
But it is important to note that in Schrodinger's conception, the wave
function for the matter field does not relate to a single quantity of
micromatter (an electron or an atom, ect). It rather relates to an entire
ensemble of matter components. The implication here is that there is no
primitive atomistic model of charged matter. Its fundamental description
is instead in terms of continuous matter fields.
[end quote---M. Sachs _Einstein Versus Bohr_ p.98]
Schrodinger discovered the wave function solution, but he never had any use
for the probabilistic interpretation of it. What else would he mean by the
cat paradox? Did he just hate cats and hope to inspire many scientists to
subject them to dangerous experiments?
Schrodinger evidently believed looking at matter as individual particles
wasn't much use. In his book Mendel Sachs goes even further arguing that
relativity and QM are mutually exclusive and that the contest should be
decided in favor of relativity which would (according to him) necessitate a
holistic view of the universe where matter is seen be made of waves instead
of particles.
-RN
Paul J. Gans
>In article <71baoi$o...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
>>Hey!!!!! We S&M types matured even earlier than any of the
>>quantum types. Of course, some might say that the field
>>ossified before quantum mechanics reached its current state,
>>but I wouldn't go *that* far.
> What I was thinking of, was that Stat Mech was the last holdout of
> the pure paper-and-pencil theorists, and many of those guys believed
> that computational theorists were an intellectually inferior breed.
> (Apparently they'd never studied Gauss, Fermi or Chandrasekhar, who'd
> been doing "computational physics" back before there _were_ computers.)
> I still detected some of this attitude when I was a grad student
> in the 1980's; it must have been much worse in the early 1960's.
> (For the benefit of others reading this thread, Paul Gans was one of
> the first SM theorists to embrace computation.)
> ------
> Robert
Thanks for saying that. I *do* go back a long way. Back
when I was a grad student at what was then the Case
Institute of Technology, there was an undergraduate
hanging around the computer center named Donald Knuth.
He was pretty sharp. I wonder what ever happened to him?
;-)
Paul J. Gans
>> In article <71baoi$o...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
>> >Hey!!!!! We S&M types matured even earlier than any of the
>> >quantum types. Of course, some might say that the field
>> >ossified before quantum mechanics reached its current state,
>> >but I wouldn't go *that* far.
>> What I was thinking of, was that Stat Mech was the last holdout of
>> the pure paper-and-pencil theorists, and many of those guys believed
>> that computational theorists were an intellectually inferior breed.
>> (Apparently they'd never studied Gauss, Fermi or Chandrasekhar, who'd
>> been doing "computational physics" back before there _were_ computers.)
>> I still detected some of this attitude when I was a grad student
>> in the 1980's; it must have been much worse in the early 1960's.
>> (For the benefit of others reading this thread, Paul Gans was one of
>> the first SM theorists to embrace computation.)
>Do S&M theorists study chemical bondage? :)
You got it. After all, chemistry is the study of the
sociology of atoms (which turns out to be rather
complicated). And many, if not most, of those atoms
are into bondage.
Just don't tell my fundie friends or they'll be out
to cut the NSF budget.
Richard Harter
Does this mean that copper atoms go around in studded leather jackets?
Are Bismuth atoms bikies?
>Just don't tell my fundie friends or they'll be out
>to cut the NSF budget.
Just move it over to the National Endowment for the Arts.
Richard Harter, c...@tiac.net, The Concord Research Institute
URL = http://www.tiac.net/users/cri, phone = 1-978-369-3911
If you can laugh at something it can't hurt you.
It can kill you but it can't hurt you.
maff91
>Robert Parson (rpa...@spotNO.SPAMColorado.edu) wrote:
>>In article <71baoi$o...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
>
>>>Hey!!!!! We S&M types matured even earlier than any of the
>>>quantum types. Of course, some might say that the field
>>>ossified before quantum mechanics reached its current state,
>>>but I wouldn't go *that* far.
>
>> What I was thinking of, was that Stat Mech was the last holdout of
>> the pure paper-and-pencil theorists, and many of those guys believed
>> that computational theorists were an intellectually inferior breed.
>> (Apparently they'd never studied Gauss, Fermi or Chandrasekhar, who'd
>> been doing "computational physics" back before there _were_ computers.)
>> I still detected some of this attitude when I was a grad student
>> in the 1980's; it must have been much worse in the early 1960's.
>> (For the benefit of others reading this thread, Paul Gans was one of
>> the first SM theorists to embrace computation.)
>
>> ------
>> Robert
>
>Thanks for saying that. I *do* go back a long way. Back
>when I was a grad student at what was then the Case
>Institute of Technology, there was an undergraduate
>hanging around the computer center named Donald Knuth.
>He was pretty sharp. I wonder what ever happened to him?
>
>;-)
"If you think you're a really good programmer, or if you want to
challenge your knowledge, read the "Art of Computer Programming," by
Donald Knuth. Be sure to solve the problems.
Knuth's work is published by Addison-Wesley in three volumes, with
more to come. The volumes are titled, "Fundamental Algorithms,"
"Semi-Numerical Algorithms," and "Sorting & Searching." If somebody is
so brash that they think they know everything, Knuth will help them
understand that the world is deep and complicated.
It took incredible discipline, and several months, for me to read it.
I studied 20 pages, put it away for a week, and came back for another
20 pages. You should definitely send me a resume if you can read the
whole thing. " - Bill Gates
http://www.microsoft.com/billgates/columns/1995q&a/qa950411.htm
>
> ----- Paul J. Gans [ga...@panix.com]
*****************************************************
"Science is the true theology" -- Thomas Paine
(as quoted in Emerson: The Mind on Fire page 153)
"The Age of Paine" by Jon Katz
http://www.wired.com/wired/archive/3.05/paine.html
*****************************************************
MJR
>
> In article <71baoi$o...@panix2.panix.com>, Paul J. Gans <ga...@panix.com> wrote:
>
> >Hey!!!!! We S&M types matured even earlier than any of the
> >quantum types. Of course, some might say that the field
> >ossified before quantum mechanics reached its current state,
> >but I wouldn't go *that* far.
>
> What I was thinking of, was that Stat Mech was the last holdout of
> the pure paper-and-pencil theorists, and many of those guys believed
> that computational theorists were an intellectually inferior breed.
> (Apparently they'd never studied Gauss, Fermi or Chandrasekhar, who'd
> been doing "computational physics" back before there _were_ computers.)
> I still detected some of this attitude when I was a grad student
> in the 1980's; it must have been much worse in the early 1960's.
> (For the benefit of others reading this thread, Paul Gans was one of
> the first SM theorists to embrace computation.)
I'd like to second this observation. As a young punk grad stud, my
stat mech professor told me to my face that if you have to use a
computer you haven't thought enough about your problem. Professional
pride not withstanding, I am not unsympathetic to the attitude: it can
be quite tempting to let the computer do the thinking, and to calculate
quantities simply because it is convenient to do so. But, one needn't
go overboard.
I seem to recall that in 'The Making of the Atom Bomb', by Richard
Rhodes, there is a paragraph about Fermi being derided as a "quantum
engineer". Evidently some people were miffed that his success was
founded upon his brilliant computational skills rather than the proper
sort of physical insight. Some things never change.
Paul J. Gans
Insight comes in many forms. Most of us have precious
little of it. Most scientists would be happy to have
*one* really great idea per career. There are a few
who have had *two*. I'm in luck, I haven't had my one
yet. ;-)
For me, numbers can give me some feeling for a problem. I stare
at them, I graph them, I combine them in various ways, and when
I am done, I often have a much better feeling for how the
physics of the situation actually works. *Then* I sit down
(actually I pace) and start thinking hard. This often leads
to another period of number generation and the entire cycle
repeats. Sometimes I even reach conclusions... ;-)
Other folks do things quite differently. I know that, for
instance, the quantum mechs I share this end of the floor
with have gone ga-ga over three-dimensional visualization
tools. We've got all sorts of strange things rotating
away on the screens of the SGI hardware infesting this
place. But they seem quite happy about it.
But I know organic chemists who can *guess* the electron
distribution in a complex molecule better than most QM
folks can calculate it. That's insight of a different
kind.
MJR
>
> For me, numbers can give me some feeling for a problem. I stare
> at them, I graph them, I combine them in various ways, and when
> I am done, I often have a much better feeling for how the
> physics of the situation actually works. *Then* I sit down
> (actually I pace) and start thinking hard. This often leads
> to another period of number generation and the entire cycle
> repeats. Sometimes I even reach conclusions... ;-)
>
> Other folks do things quite differently. I know that, for
> instance, the quantum mechs I share this end of the floor
> with have gone ga-ga over three-dimensional visualization
> tools. We've got all sorts of strange things rotating
> away on the screens of the SGI hardware infesting this
> place. But they seem quite happy about it.
As a fanatical disciple of Neil Postman, of NYU by coincidence, I am
somewhat troubled by this trend in my field. My guru documents the way
shifts in the representation of ideas introduce subtle and profound
changes in the nature of what is thought. The visualisation software is
incredibly seductive in the way it brings molecules to life but it
highlights differenct aspects of the chemistry from good old pencil &
paper analysis of the calculations; perhaps in the end this will be an
unalloyed Good Thing, but I reserve the right to suspend judgement until
all the precincts are heard from.
> But I know organic chemists who can *guess* the electron
> distribution in a complex molecule better than most QM
> folks can calculate it. That's insight of a different
> kind.
>
> ----- Paul J. Gans [ga...@panix.com]
I stand in awe of the ability of my experimental colleagues, and I
live for the day when-finally-I can prove that one of them has guessed
wrong. <:p
Rex Nielson
> In article <rnielson-ya0240800...@news.xmission.com>,
rnie...@xmission.com (Rex Nielson) wrote:
>
> > I may as well come clean, I have read a book about physics. In fact I've
> > been drawing heavily from theoretical physicist Mendel Sachs' book for
> > non-experts _Einstein Versus Bohr_ .
>
> Oh, great. Try an intro QM text instead. Maybe vol. III of the Feynman
> Lectures, the intro stuff is very good without too much math.
>
> > In which he makes a point of
> > emphasizing the distinction between subjectivity and objectivity. He says
> > "irreducible subjectivity in the role of measuring apparatus as a
> > fundamental ingredient in our understanding of matter"
>
> That doesn't sound like a physics book to me. That sounds like a book on
> philosophy. Unsurprising, given Sach's Ph.D. in "philosophy of physics".
> (Perhaps the discussion of "pluralism", "monism", "logical positivism",
> "realism", "irreducible subjectivity", etc. might have tipped you off.)
Why should familiarity with philosophical terminology count against him?
He has researched relativity and QM for over three and a half decades, I
have no doubt he understands the concepts involved including the
mathematical ones---or at least his book was full of a lot of formulas I
didn't understand.
>
> > He negatively contrasts this with the
> > fundamentally "objective" theory of relativity. He says for this and other
> > reasons QM and the theory of relativity are incompatible.
>
> They certainly aren't "incompatible". At least not QM and _special_
> relativity. The problems reconciling QM and _general_ relativity are
> well-known.
Then why is his position so controversial to you? Since our universe is
full of non-uniform motion, it's phenomena is better described by general
relativity then special relativity. If the fact that general relativity
doesn't reconcile with quantum mechanics is well known, then the substance
of Sachs' argument seems pretty uncontroversial. You seem to be admitting
there are problems between QM and general relativity. Why should these
problems be hidden and ignored, and how will they ever be resolved if they
are?
> > This would seem to put modern physics in quite a pickle, don't you think?
>
> You can't put a physical theory in a "pickle" by philosophy. Sachs can
> argue all he wants that "irreducible subjectivity" blah blah blah is a
> "fundamental ingredient" for a physical theory, but that does not prove
> that QM and relativity are fundamentally incompatible. To do that you
> need a mathematical proof.
>
> He also claims that relativistic quantum field theory is neither
> mathematically nor logically consistent, which is an unjustifiable claim.
> It simply is not known whether or not RQFT is a consistent theory.
> It _is_ known that the predictions of RQFT agree with experiment to
> unprecedented levels of precision. (Superstring theories also get rid
> of some the mathematical problems of QFTs that Sachs was complaining
> about, and they certainly aren't incompatible with relativity..) It may
> be that relativity and QM are _not_ mutually compatible, but I didn't
> see anything on his page even remotely implying that he had a _proof_,
> just a bunch of philosophical hand-waving.
He says RQFT has no solutions because of automatically generated
infinities. You just said, "the problems reconciling QM and _general_
relativity are
well-known." It's the compatibility of *general* relativity that's at
stake because that's the form of relativity that's supposed to describe the
real world. If I've got it right, for at least one reason (an implication
of instantaneous action-at-a-distance) Sachs would say that QM is
incompatible with both special and general relativity, but reconciling QM
and *general* relativity would be the most important, and you concede there
are difficulties with this.
> I should point out that you have chosen to learn your quantum mechanics
> and relativity from someone who is far out on the fringe of the scientific
> community. YMMV.
I question how you judge who is on the fringe. My rhetoric shouldn't be
confused with the opinions of Sachs. He presents a seemingly reasonable
proposal that this acknowledged difficulty between QM and relativity may
have implications for the future of physics and should be considered. Why
is it so radical to say that the current conventions of today's physics,
that replaced the conventions of yesterday's physics, may themselves be due
for an overhaul. That seems a lot more reasonable then saying our universe
may have been given birth by a black hole in an other universe. Is
discussing the implications of apparently well known problems in physics
worse science then the type of speculation you yourself admit is weird?
-RN
"Indeed the finite interaction between object and measuring agencies
conditioned by the very existence of the quantum action entails... the
necessity of a final renunciation of the classical idea of causality and a
radical revision of our attitude toward the problem of physical reality."
-Niels Bohr
Michael L. Siemon
+ Why should familiarity with philosophical terminology count against him?
+ He has researched relativity and QM for over three and a half decades, I
+ have no doubt he understands the concepts involved including the
+ mathematical ones---or at least his book was full of a lot of formulas I
+ didn't understand.
Oh my; he has "lots of formulas" that you "didn't understand" and on *this*
basis you think he is reliable? I suggest that in the circumstances, you
would be better served by *having* doubts (instead of "no doubt") about
his other statments (which you may or may not actually have understood,
since you have no way of knowing what relation they have to the math
you specifically disclaim understanding of.)
--
Michael L. Siemon We must know the truth, and we must
m...@panix.com love the truth we know, and we must
act according to the measure of our love.
-- Thomas Merton
tpham...@my-dejanews.com
Thanks! On further reading, my guess that the BEC was not coherent
was wrong.
> > > How can matter be a
> > > wave and a particle at the same time...
> > That is why I prefer to think of atoms as particles, and the wave
> > nature as just "quantum behavior".
> Isn't "quantum behavior" often seen as pretty topsy-turvy? At the quantum
> level things don't seem to necessarily follow cause and effect
> relations---electrons unpredictably jumping to lower energy states, photons
> arbitrarily being created and destroyed. Objects don't suddenly change, or
> blink in and out of existence in the larger world we know and live in.
> Wouldn't the appearance of material wave phenomena on the macroscopic scale
> we are familiar with call into question the assumption that the wave nature
> of matter can be attributed to "quantum behavior"?
Nope. If you get coherence between a large number of particles, then
macroscopic quantum phenomena are observable. I should have kept
this in mind when I made my guess (a source I quickly consulted
said nothing about the coherence), because the only way you will
see macroscopic quantum behavior is if the converse is true. Large
objects behave classically because they are *non-coherent* states.
Objects blinking in and out of existence is determined by the Heisenberg
Uncertainty principle, which quite effectively removes the probability
of large objects (atoms even) from virtual particlehood.
> > > The answer QM gave was that 'pure'
> > > matter was shrouded by its smallness and the very act of looking at it
> > > would affect its appearance,
> > It is very important to note that this is true *even if the act
> > of looking at the particle does not physically disturb the system.*
> > I would say it affects its wave function rather than its appearance.
> > > so it was as far as we were concerned it could
> > > be both wave and particle.
> > The concept is not necessary, and I maintain it makes no sense.
> I was attempting to say that according to QM, matter is attributed to
> particles, even though particles can't be directly seen. The reason
> particles can't be directly observed is attributed fundamental limitations
> on our ability to measure. This would necessitate a probabilistic
> interpretation.
The concept of a particle is just that - something used to explain
nature. The term direct observation is vague. Is imaging by
scanning tunelling microscope or atomic force microscope direct
observation? How about x-ray crystallography, electron and
neutron diffraction? Just what is probabilistic about the
interpretations of these experiments?
The concept of matter predates QM by a long shot, and QM actually
quite explicitly uses the concept as already established. QM uses the
existence of something we named particles that behave with particle behavior.
For example, each term in a molecular Hamiltonian is for an
electron or nucleus, (absent any other interactions besides
coulombic).
> > > QM's probabilistic *subjective* interpretation
> > > of matter is legitimated by this indeterminacy. But is this a valid
> > > inquiry into the true nature of matter? With *objective* wave phenomena
in
> > > matter now being observed it would seem necessary to call QM's
presumptions
> > > into question. Perhaps QM is not a description of reality only a handy
> > > mathematical approximation.
> > Parts of this are incoherent. I find it odd that the BEC is a prediction
from
> > quantum phenomena, yet you say that the BEC is a reason to call quantum
> > theory into question. The issues of interpretation are quite separate
> > from whether QM is a description of reality (from the computational
> > point of view).
> > Perhaps you could be a little clearer on your point. Is it
> > that matter is not a particle, but just a wave? Just because it
> > has quantum behavior is not sufficient, because the waves are
> > probability functions.
> All the articles I've read about the BEC describe the it's unique
> attributes as QM effects writ large. Wave behavior in matter is
> automatically associated with Quantum Mechanics, but this seems to be
> merely convention at work because probabilistic activity of particles isn't
> the only way interpret wave behavior in matter. According to what I've
> read conventional QM wasn't even the original way to interpret wave
> behavior in matter:
> [quote]
> But it is important to note that in Schrodinger's conception, the wave
> function for the matter field does not relate to a single quantity of
> micromatter (an electron or an atom, ect). It rather relates to an entire
> ensemble of matter components. The implication here is that there is no
> primitive atomistic model of charged matter. Its fundamental description
> is instead in terms of continuous matter fields.
> [end quote---M. Sachs _Einstein Versus Bohr_ p.98]
What is a "matter field"? It sounds like hooey to me.
If there is no atomistic model of charged matter, then
why is there a very well defined term for each particle in the QM Hamiltonian?
> Schrodinger discovered the wave function solution, but he never had any use
> for the probabilistic interpretation of it. What else would he mean by the
> cat paradox? Did he just hate cats and hope to inspire many scientists to
> subject them to dangerous experiments?
Er, I think the point of Schrodinger's cat was that certain probabilistic
QM interpretations are weird. For example, what would the wave function
mean for half dead half alive cat? The analogous things for particles come up
frequently. One way around this is to say the wave function is a
description of an ensemble where the probabilities of dead or alive are
one half. With particles, you have an ensemble of identically
prepared states, however each system is a particle system.
> Schrodinger evidently believed looking at matter as individual particles
> wasn't much use. In his book Mendel Sachs goes even further arguing that
> relativity and QM are mutually exclusive and that the contest should be
> decided in favor of relativity which would (according to him) necessitate a
> holistic view of the universe where matter is seen be made of waves instead
> of particles.
I don't see how visualizing matter as waves is useful. Either you or Sachs
need to be more explicit here.
Rex Nielson
If the wave properties of BEC's are seen as a coherent state between a
large number of particles, then clearly the wave properties of matter
aren't always subject to a probabilistic interpretation. In crystal
diffraction experiments, where in effect a stream of electrons is projected
onto a screen, the probabilistic interpretation is at least conceivable by
imagining that the observed wave properties are merely a result of the
arbitrary trajectories of individual electrons whose resulting impacts
could be described by probability rules. Since with BEC experiments wave
phenomena in matter must be described as "coherence between a large number
of particles" then, in contrast with the earlier experiments, matter-waves
cannot be attributed to the probabilistic behavior of individual particles.
Earlier on this thread you said:
"the "wave" in the wave nature is a *probability* amplitude. This is not
the same as a wave in water or other medium"
It would seem that this statement, and the commonplace interpretation of
wave/particle duality that it reflects, must be wrong. The BEC wave is
compared to a liquid, they prod it, adjust it, shine light through it,
manipulate it into an 'atom laser.' Thus wave features are shown to be a
totally objective and native property of elemental matter when it's cold
enough, its waves *are* comparable those of water and other mediums at more
familiar temperatures. The wave nature of matter is not a "probability
amplitude" unless the laws of probability change in very cold conditions
and something impossible at regular temperatures---the concerted
oscillation of a large number of atoms---becomes somehow probable. I don't
think the laws of probability are subject to temperature change but I've
put some lottery tickets in the freezer just in case.
-RN
>
> > > > The answer QM gave was that 'pure'
> > > > matter was shrouded by its smallness and the very act of looking at it
> > > > would affect its appearance,
>
> > > It is very important to note that this is true *even if the act
> > > of looking at the particle does not physically disturb the system.*
> > > I would say it affects its wave function rather than its appearance.
>
> > > > so it was as far as we were concerned it could
> > > > be both wave and particle.
>
> > > The concept is not necessary, and I maintain it makes no sense.
>
> > I was attempting to say that according to QM, matter is attributed to
> > particles, even though particles can't be directly seen. The reason
> > particles can't be directly observed is attributed fundamental limitations
> > on our ability to measure. This would necessitate a probabilistic
> > interpretation.
>
> The concept of a particle is just that - something used to explain
> nature. The term direct observation is vague. Is imaging by
> scanning tunelling microscope or atomic force microscope direct
> observation? How about x-ray crystallography, electron and
> neutron diffraction? Just what is probabilistic about the
> interpretations of these experiments?
Obviously particles at larger levels can be seen,I can see them myself
(dust particles at least) accumulating on my window sill. But according to
the Heisenberg Uncertainty principle the resolution of elementary parts of
matter is fundamentally limited, so matter at the smallest level is held
unaccountable for the strange behavior attributed to it.
>
> The concept of matter predates QM by a long shot, and QM actually
> quite explicitly uses the concept as already established. QM uses the
> existence of something we named particles that behave with particle behavior.
> For example, each term in a molecular Hamiltonian is for an
> electron or nucleus, (absent any other interactions besides
> coulombic).
Yes, atomism is as at least as old as Democritus.
tpham...@my-dejanews.com
rnie...@xmission.com (Rex Nielson) wrote:
> In article <71koe0$1ba$1...@nnrp1.dejanews.com>, tpham...@my-dejanews.com
wrote:
>
> > In article <rnielson-ya0240800...@news.xmission.com>,
> > rnie...@xmission.com (Rex Nielson) wrote:
> > > In article <71codl$t95$1...@nnrp1.dejanews.com>, tpham...@my-dejanews.com
> > wrote:
> > > > If I may interrupt, the "wave" in the wave nature is a *probability*
> > > > amplitude. This is not the same as a wave in water or other medium.
> > > > We say there is wave behavior *if there is interference*. The wave
> > > > function must also satisfy certain symmetry requirements, depending
> > > > on whether they are bosons or fermions.
[snip]
Boy, are you confused! The atoms each have a wave function. Coherence
is constructive interference between each wave function. The
interpretation of the wave function is that it gives the probability
of finding the particle in that region of space.
The experiment showed the BEC was a coherent superposition, because two
BEC coherent states show a well defined interference pattern of
constructive and destructive interference. The constructive interference
and destructive interference, when interpreted as probabilities of
finding the partcles or not, gives areas where no atoms will be seen, and
areas where they will be found. This was what was seen in the experiment,
which substantiates this conventional interpretation.
Why is the BEC seen only at extremely low temperatures? Because it depends
on all atoms being in the ground state. If there is heat, then there will
be thermal excitation. This changes the state and therefore the wave
function, and therefore the probability density. So in this way, probability
does depend on temperature. This is trivially seen by the fact that
BECs only exist at extremely low temperatures.
By the way, you can take the lottery tickets out of the freezer. You
are more likely to win if you raise the temperature to 451F.
[snip]
> > The concept of a particle is just that - something used to explain
> > nature. The term direct observation is vague. Is imaging by
> > scanning tunelling microscope or atomic force microscope direct
> > observation? How about x-ray crystallography, electron and
> > neutron diffraction? Just what is probabilistic about the
> > interpretations of these experiments?
> Obviously particles at larger levels can be seen,I can see them myself
> (dust particles at least) accumulating on my window sill.
That is why the particle model is so useful. We understand what
a particle is.
> But according to
> the Heisenberg Uncertainty principle the resolution of elementary parts of
> matter is fundamentally limited, so matter at the smallest level is held
> unaccountable for the strange behavior attributed to it.
The Heisenberg Uncertainty principle only limits the variance of
variables for which the operators do not commute. I am not sure what
holding matter accountable for strange behavior would be -
perhaps trials in tiny little courts?
> > The concept of matter predates QM by a long shot, and QM actually
> > quite explicitly uses the concept as already established. QM uses the
> > existence of something we named particles that behave with particle
behavior.
> > For example, each term in a molecular Hamiltonian is for an
> > electron or nucleus, (absent any other interactions besides
> > coulombic).
> Yes, atomism is as at least as old as Democritus.
I noticed you don't explain why particles are explicitly included in QM
if "matter is waves".
[snip]
> > What is a "matter field"? It sounds like hooey to me.
> > If there is no atomistic model of charged matter, then
> > why is there a very well defined term for each particle in the QM
Hamiltonian?
Well?
[snip]
Rex Nielson
Notice how you say that *each* atom has a wave function. You define wave
function as the probability of finding *the* particle in *that* region of
space. Then without missing a beat you say that interference in BECs
substantiates the conventional interpretation. Remember a BEC is supposed
to a vast conglomeration of atoms in a region of space large enough to see
with the naked eye. How can the wave function be the probability of
finding a particle in a very limited region of space and also be the
probability of finding a large number of particles in a large region of
space. It's as if you were using statistical analysis to convince a
gambler that his slot machine was behaving in a random manner when suddenly
every slot-machine in the casino begins to simultaneously get the same
results. All the slot machines would now be subject to the same analysis
would this be even more proof that their behavior was random?
>
> Why is the BEC seen only at extremely low temperatures? Because it depends
> on all atoms being in the ground state. If there is heat, then there will
> be thermal excitation. This changes the state and therefore the wave
> function, and therefore the probability density. So in this way, probability
> does depend on temperature. This is trivially seen by the fact that
> BECs only exist at extremely low temperatures.
What is the wave function of elemental matter? If it's merely the
probability of finding a particle in a certain region of space then
shouldn't the probabilities for finding these particles change as
conditions changed the proportions of space in which you would be looking
for something? If wave function is the mere probabilistic result of some
mechanical action why would it be similar in crystal diffraction and BEC
experiments where the things measured are in totally different proportions
under totally different conditions?
>
> By the way, you can take the lottery tickets out of the freezer. You
> are more likely to win if you raise the temperature to 451F.
Curious, that's exactly the temperature the collected works of Ray Bradbury
burned at.
>
> > Obviously particles at larger levels can be seen,I can see them myself
> > (dust particles at least) accumulating on my window sill.
>
> That is why the particle model is so useful. We understand what
> a particle is.
>
> > But according to
> > the Heisenberg Uncertainty principle the resolution of elementary parts of
> > matter is fundamentally limited, so matter at the smallest level is held
> > unaccountable for the strange behavior attributed to it.
>
> The Heisenberg Uncertainty principle only limits the variance of
> variables for which the operators do not commute. I am not sure what
> holding matter accountable for strange behavior would be -
> perhaps trials in tiny little courts?
Particle court's first case on the docket: The state vs. 2000 rubidium
atoms. These particles are accused of losing their individuality and
lasciviously, and illegally sharing all behavior when cooled to 100
billionth of a degree above absolute zero. By exhibiting wave properties
at macroscopic scale and indecently revealing their true nature to innocent
scientists they are charged with violating the Heisenberg Uncertainty
principle.
>
> > > The concept of matter predates QM by a long shot, and QM actually
> > > quite explicitly uses the concept as already established. QM uses the
> > > existence of something we named particles that behave with particle
> behavior.
> > > For example, each term in a molecular Hamiltonian is for an
> > > electron or nucleus, (absent any other interactions besides
> > > coulombic).
>
> > Yes, atomism is as at least as old as Democritus.
>
> I noticed you don't explain why particles are explicitly included in QM
> if "matter is waves".
They as a concept are undeniably handy, one that I'm not ready to go without.
>
> [snip]
>
> > > What is a "matter field"? It sounds like hooey to me.
> > > If there is no atomistic model of charged matter, then
> > > why is there a very well defined term for each particle in the QM
> Hamiltonian?
>
> Well?
>
According to Sach's, relativity can't have singularities. Also he thinks
that for complete objectivity from a relative view, the inertial frame of
the observer would have to be taken into account. For this to be done
manifestations of mass would have to be integrated into a complete field
theory so a piece of matter would make about as much sense as a piece of
gravity. That description may not be right on because I've got to hurry to
work and don't have time to look up quotes. Anyway, when he talks about
the groundwork he claims to have laid for a new theory he mostly just
refers to his more technical work. I'm not prepared to explain away every
apparent manifestation of matter's particle nature, I wouldn't even know
where to start, but I think his work warrants consideration because the
paradox of wave/particle duality seems to be growing more pronounced.
-RN
tpham...@my-dejanews.com
[snip]
> > Boy, are you confused! The atoms each have a wave function. Coherence
> > is constructive interference between each wave function. The
> > interpretation of the wave function is that it gives the probability
> > of finding the particle in that region of space.
> >
> > The experiment showed the BEC was a coherent superposition, because two
> > BEC coherent states show a well defined interference pattern of
> > constructive and destructive interference. The constructive interference
> > and destructive interference, when interpreted as probabilities of
> > finding the partcles or not, gives areas where no atoms will be seen, and
> > areas where they will be found. This was what was seen in the experiment,
> > which substantiates this conventional interpretation.
>
> Notice how you say that *each* atom has a wave function. You define wave
> function as the probability of finding *the* particle in *that* region of
> space. Then without missing a beat you say that interference in BECs
> substantiates the conventional interpretation. Remember a BEC is supposed
> to a vast conglomeration of atoms in a region of space large enough to see
> with the naked eye. How can the wave function be the probability of
> finding a particle in a very limited region of space
who said limited? The probability amplitudes are over ALL space, though
as a practical matter, they are most likely to be found where they
are usually found. :) The BEC wave function is a coherent *superposition*
of wave functions of individual particles. That is what makes it
behave in its special way.
The BEC wave function gives the probability distribution of each particle.
> and also be the
> probability of finding a large number of particles in a large region of
> space. It's as if you were using statistical analysis to convince a
> gambler that his slot machine was behaving in a random manner when suddenly
> every slot-machine in the casino begins to simultaneously get the same
> results. All the slot machines would now be subject to the same analysis
> would this be even more proof that their behavior was random?
Your analogy would mean something if slot machines had interference effects.,
and if matter behaved in a random manner. An H atom has a probabilistic
distribution of electron density, yet well defined spectroscopic
properties. Does it behave randomly? I say it is an ill-posed question.
> > Why is the BEC seen only at extremely low temperatures? Because it depends
> > on all atoms being in the ground state. If there is heat, then there will
> > be thermal excitation. This changes the state and therefore the wave
> > function, and therefore the probability density. So in this way,
probability
> > does depend on temperature. This is trivially seen by the fact that
> > BECs only exist at extremely low temperatures.
> What is the wave function of elemental matter?
It depends on its state, which depends on its potential! The BEC is
a collection of ATOMS, so the relevant wave function is the ground
state of the atoms, which are well known. The wave function of quarks
etc is not needed at the level of atoms.
> If it's merely the
> probability of finding a particle in a certain region of space then
> shouldn't the probabilities for finding these particles change as
> conditions changed the proportions of space in which you would be looking
> for something?
A confinement changes the potential, which influences the wave function,
which gives the probability distribution. Surely you have heard of
the particle in a box. However, absent any physical effect, the
probability distribution is not affected by the fact that you are only
"interested" in a certain region.
> If wave function is the mere probabilistic result of some
> mechanical action why would it be similar in crystal diffraction and BEC
> experiments where the things measured are in totally different proportions
> under totally different conditions?
Because the wave functions are very similar types, presumably.
> > By the way, you can take the lottery tickets out of the freezer. You
> > are more likely to win if you raise the temperature to 451F.
> Curious, that's exactly the temperature the collected works of Ray Bradbury
> burned at.
That must just be a random number I picked out then. :)
> > > Obviously particles at larger levels can be seen,I can see them myself
> > > (dust particles at least) accumulating on my window sill.
> >
> > That is why the particle model is so useful. We understand what
> > a particle is.
> >
> > > But according to
> > > the Heisenberg Uncertainty principle the resolution of elementary parts of
> > > matter is fundamentally limited, so matter at the smallest level is held
> > > unaccountable for the strange behavior attributed to it.
> >
> > The Heisenberg Uncertainty principle only limits the variance of
> > variables for which the operators do not commute. I am not sure what
> > holding matter accountable for strange behavior would be -
> > perhaps trials in tiny little courts?
>
> Particle court's first case on the docket: The state vs. 2000 rubidium
> atoms. These particles are accused of losing their individuality and
Hmmm. That's interesting. The bosons should be indistinguishable,
but...
> lasciviously, and illegally sharing all behavior when cooled to 100
> billionth of a degree above absolute zero. By exhibiting wave properties
> at macroscopic scale and indecently revealing their true nature to innocent
> scientists they are charged with violating the Heisenberg Uncertainty
> principle.
Sharing behavior is not defined under the laws of QM. And showing
macroscopic wave behavior is not a violation of the HUP.
Case dismissed.
> > > > The concept of matter predates QM by a long shot, and QM actually
> > > > quite explicitly uses the concept as already established. QM uses the
> > > > existence of something we named particles that behave with particle
> > behavior.
> > > > For example, each term in a molecular Hamiltonian is for an
> > > > electron or nucleus, (absent any other interactions besides
> > > > coulombic).
> >
> > > Yes, atomism is as at least as old as Democritus.
> >
> > I noticed you don't explain why particles are explicitly included in QM
> > if "matter is waves".
>
> They as a concept are undeniably handy, one that I'm not ready to go without.
Good. Same here and for most scientists.
[snip]
> According to Sach's, relativity can't have singularities. Also he thinks
> that for complete objectivity from a relative view, the inertial frame of
> the observer would have to be taken into account. For this to be done
> manifestations of mass would have to be integrated into a complete field
> theory so a piece of matter would make about as much sense as a piece of
> gravity. That description may not be right on because I've got to hurry to
> work and don't have time to look up quotes. Anyway, when he talks about
> the groundwork he claims to have laid for a new theory he mostly just
> refers to his more technical work. I'm not prepared to explain away every
> apparent manifestation of matter's particle nature, I wouldn't even know
> where to start, but I think his work warrants consideration because the
> paradox of wave/particle duality seems to be growing more pronounced.
Many of us don't see it as a paradox. QM may just be incomprehensible
(maybe forever). Just because some physicist mayhave some ideas means
little at this point. If Sachs manages to show reason for considering
matter as not being particulate, I am sure the physics community will
be all ears.
Tracy P. Hamilton (temporarily sigless. For only pennies a day, Sally
Struthers can see to it that I get one. You will receive a photo and
a letter from me.)
Rex Nielson
> In article <rnielson-ya0240800...@news.xmission.com>,
> rnie...@xmission.com (Rex Nielson) wrote:
> > Notice how you say that *each* atom has a wave function. You define wave
> > function as the probability of finding *the* particle in *that* region of
> > space. Then without missing a beat you say that interference in BECs
> > substantiates the conventional interpretation. Remember a BEC is supposed
> > to a vast conglomeration of atoms in a region of space large enough to see
> > with the naked eye. How can the wave function be the probability of
> > finding a particle in a very limited region of space
>
> who said limited? The probability amplitudes are over ALL space, though
> as a practical matter, they are most likely to be found where they
> are usually found. :) The BEC wave function is a coherent *superposition*
> of wave functions of individual particles. That is what makes it
> behave in its special way.
>
> The BEC wave function gives the probability distribution of each particle.
Could a probability amplitude exist over all space? Yes, because it's not
a thing, it's just a range of possibilities. Could a wave exist across all
space? Like a ripple extending across a pond, it would be a very natural
thing for a wave. On the other it would seem to defy the definition of the
particle as a localized entity to have it smeared across the universe. So
if wave function extends through all space then it must describe either a
probability amplitude (of where a particle might be) or an actual wave.
What would most distinguish a probability amplitude from a wave? One is an
abstract representation of the possibility that something might happen and
the other is an observable physical entity. By all accounts the BEC is an
actual physical entity and it's wave features are an objective property and
not a statistical illusion. A probability is nothing more then a
mathematical abstraction while the wave features of the BEC seem to be, in
every sense of the word, real. You may as well tell a surfer the wave he
is riding on isn't real, with objective, dynamic qualities of its own, but
merely a probability distribution. Since the wave function of the BEC
describes an actual wave---something real and that can be seen and touched
as opposed to something mathematical that can only be seen with the
imagination---and wave function describes something that, unlike a
particle, can exist across all space, then the wave function must describe
real wave properties in matter that underlie its particulate
manifestations.
-RN
tpham...@my-dejanews.com
rnie...@xmission.com (Rex Nielson) wrote:
> In article <7250sp$od5$1...@nnrp1.dejanews.com>, tpham...@my-dejanews.com
wrote:
>
> > In article <rnielson-ya0240800...@news.xmission.com>,
> > rnie...@xmission.com (Rex Nielson) wrote:
>
> > > Notice how you say that *each* atom has a wave function. You define wave
> > > function as the probability of finding *the* particle in *that* region of
> > > space. Then without missing a beat you say that interference in BECs
> > > substantiates the conventional interpretation. Remember a BEC is supposed
> > > to a vast conglomeration of atoms in a region of space large enough to see
> > > with the naked eye. How can the wave function be the probability of
> > > finding a particle in a very limited region of space
> >
> > who said limited? The probability amplitudes are over ALL space, though
> > as a practical matter, they are most likely to be found where they
> > are usually found. :) The BEC wave function is a coherent *superposition*
> > of wave functions of individual particles. That is what makes it
> > behave in its special way.
> >
> > The BEC wave function gives the probability distribution of each particle.
>
> Could a probability amplitude exist over all space? Yes, because it's not
> a thing, it's just a range of possibilities. Could a wave exist across all
> space? Like a ripple extending across a pond, it would be a very natural
> thing for a wave. On the other it would seem to defy the definition of the
> particle as a localized entity to have it smeared across the universe.
The particle is not smeared out. The probability distribution is.
So the particle is likely to be found anywhere. Some places more likely
than others.
> So
> if wave function extends through all space then it must describe either a
> probability amplitude (of where a particle might be) or an actual wave.
> What would most distinguish a probability amplitude from a wave? One is an
> abstract representation of the possibility that something might happen
with observable consequences!
> and
> the other is an observable physical entity. By all accounts the BEC is an
> actual physical entity and it's wave features are an objective property
if you want to call the observable consequences of a system with
a given wave function an "objective property", then the BEC has it.
> and
> not a statistical illusion.
The observable consequences are not an illusion. Here is the situation:
The wave function gives the probabilities of finding particles in
certain positions. An experiment is run, and the particles are
found distributed as predicted. It is not necessary to say that
particles "are" waves.
> A probability is nothing more then a
> mathematical abstraction
Really? The fact that the probability of getting heads = one half
nothing more than a mathematical abstraction?
> while the wave features of the BEC seem to be, in
> every sense of the word, real.
The probability distribution gives the probability of
making various measurements. The measurements are real,
and the probability distribution is real.
> You may as well tell a surfer the wave he
> is riding on isn't real, with objective, dynamic qualities of its own, but
> merely a probability distribution.
One of my points is that the "wave" in QM is the probability
amplitude, NOT the same as a wave of water. They are both
real but *different* things. A water wave is not a coherent
quantum state.
> Since the wave function of the BEC
> describes an actual wave---something real and that can be seen and touched
There is no prohibition that coherent states can't produce
something that behaves the same as an actual wave. Is that
the source of your confusion?
> as opposed to something mathematical that can only be seen with the
> imagination
And that predicts the energy levels of atoms very precisely,
and the energy levels of molecules from linear combination
of molecular orbitals, molecular structures, and more
unobservable and imaginary stuff like that.
> ---and wave function describes something that, unlike a
> particle, can exist across all space,
Can exist at any *point* in space. There's a difference.
> then the wave function must describe
> real wave properties in matter that underlie its particulate
> manifestations.
I don't see why it can't. It explains the BEC, which you claim is
a "real wave property". But the wave properties in BECs are not
under the particulate nature. The particles are just particles,
the *particles* have wave functions, the wave functions interfere
(a quite mysterious process), and the interference pattern is observed.
Just what is being observed? Where the particles are, which is determined by
what? Their probability distribution. NO problem for the standard
interpretation.
-----------------------------------------------------------------------
Tracy P. Hamilton |Have you ever noticed? Anybody going slower
Building Manager, Alco Hall |than you is an idiot, and anyone going
University of Ediacara |faster than you is a maniac. * George Carlin
-----------------------------------------------------------------------
-----------== Posted via Deja News, The Discussion Network ==----------