Friday, January 29, 2016

All university managers should have to teach


When I was a graduate student at Princeton it was known that the President, William Bowen, regularly taught an undergraduate class. I recall reading that he thought that this was important so he did not lose touch with what the institution was all about.
Much later in 2011 Bowen also said

"Teaching and doing research are very good preparation for this kind of job because you have to analyze issues and understand them, and you have to be able to communicate," he said. "Teaching is a great way to hone whatever skills one has in that area. A lot of being president is about teaching."

Also, while watching The Ivory Tower I noticed that Michael Roth, the president of Wesleyan College, seemed to be teaching undergraduates. Indeed, his web page says,

He continues to teach undergraduate courses, and through Coursera has offered MOOCs, the most recent being “How to Change the World.”

Does anyone know of other examples?

I think Bowen and Roth should not be anomalies and curiosities but the norm.

In fact, I think all university managers (Deans, Vice Presidents, ....) should have to do a small amount of undergraduate teaching each year. I would suggest a minimum of one half to one third of a course for second or third year students. This would be a relatively small time investment that would bring significant benefits and make them much better managers.

Why is this a desirable policy?

Managers will be more likely to formulate realistic and helpful policies.

A reminder of a core mission of the university: teaching.
Admin, fund raising, rankings, metrics, .... are just a means to an end... and sometimes an obstacle to the real mission. Furthermore, sports teams, hospitals, industry research centres, school outreach, are a distraction, whatever their merits...

It may weed out senior managers who really don't like teaching and/or students.

It will give administrators credibility when they try to persuade faculty and students of new policies and procedures, particularly ones that may be disruptive or unpopular.

What might they learn?

Teaching is hard work.
Anyone can stand up in front of a class and drone on, but actually engaging students and getting them to learn something is not easy or simple.

Students are diverse.
Even at a university such as UQ where the students are pretty uniform from the point of view of academic background (top 10-20% of high school graduates) and socio-economic background (middle class), there is a striking diversity of values, work ethic, ambitions, and ability.

Students are human beings not abstract entities.
Students are not numbers, customers, or clients. They have feelings, aspirations, frustrations, ... One learns and understands these by personal interaction not by online surveys.

Student evaluations of teaching and courses are blunt instruments with many flaws.

New technology and teaching innovations have some value but are over-hyped.

Aside. I feel I can "put my money where mouth" here. For the past 13 years I was a Research Professor and not obligated to do any teaching. However, every year on a voluntary basis I taught 0.5-1.5 courses. My perspective on students and teaching is probably quite different to if I had not done any of this teaching.

What do you think about this proposal?

Wednesday, January 27, 2016

Dynamical tunneling and overtone spectroscopy

In a molecule one can observe excitation by infra-red photons of overtones of vibrational modes, i.e. if nu is the fundamental vibrational frequency, the absorption of photons with frequency of about 2 nu, 3 nu, ... can be observed. The quantum picture is below for a 4nu absorption.



I recently learnt that overtone absorption is classically forbidden, i.e. it is intrinsically quantum mechanical (just like tunnelling, reflection above a barrier, interference, entanglement, ...). It does not occur in the limit that Planck's constant goes to zero.
Explicitly if you take an anharmonic oscillator and drive it with an external field of frequency 2 nu, you cannot get the oscillator to go at 2 nu.
Furthermore, this involves dynamical tunnelling, i.e. there is no potential barrier in real space, but rather tunnelling occurs in phase space.

There is a nice article by Eric Heller where he shows that overtone excitation is like reflection above a potential barrier.


The figure below shows the classical phase space (and Poincare surfaces) for an anharmonic (Morse) oscillator coupled to a driving field with 4 times the harmonic frequency of the oscillator.
Heller states "the local phase space structure near the [resonance] islands is the same as the above-barrier problem". See the Figure below.

Works by Lehmann and by Medvedev, explicitly shows how the transition probability for overtone excitation (i.e. the relevant matrix element) is dominated by the semi-classical dynamics in the classically forbidden region of the potential, particularly the inner wall.

This is currently of interest to me because I am working on a paper about the intensity of overtone modes in hydrogen bonded systems. In the Condon approximation overtone excitation can only occur if the potential is anharmonic. Alternatively it can arise due to non-linear terms in the dipole surface (i.e. electrical anharmonicity) (equivalent to the break-down of the Condon approximation). It does seem that the matrix element for the overtone intensity is quite sensitive to the finer details of the potential and thus the nuclear wave functions.

Tuesday, January 26, 2016

An important but basic skill: how to quickly "read" a scientific paper

Basically, look at the figures.

The amount of literature we might read is increasing exponentially. It is overwhelming. We all need some strong filters to focus on a few papers. This can save a lot of time.

The question is really, "Should I read this particular paper?"
This means answering two questions.
1. Does the paper contain some results that are of interest to me personally?
2. Are the results valid and important?

These days I tend to only look at papers that someone else recommends to me or are cited or linked to in papers I have decided to "read" using the procedure below.

The quickest and most efficient way to answer these questions is.

a. Read the title and abstract.
Is there potentially something of concrete interest to me?
If not, ditch the paper.

b. Look at the figures.
Are they comprehensible? If not, ditch the paper.
Do they contain new results, I did not know about? Are they interesting and important?
Are they valid? Do they make sense in certain limits I already know about? Are they consistent with other work?

If I get positive answers, only then will I actually look at details in the paper, such as the methods used. Then I may actually read the paper properly, perhaps even trying to work through some details.

Many senior scientists follow similar procedures.

Corollary. There are important implications of this for you when you write a paper.
Pick your title carefully.
Polish your abstract.
Work particularly hard on your figures and their captions.
Indeed, my mentor John Wilkins taught me to "write" a PRL by first constructing polished figures and captions, and then writing the text. The abstract is written last.
If you don't the paper may never get read, even though it does contain important and interesting results.


Monday, January 25, 2016

The tragic comedy of the physics job market

Peter Woit refers to this video. It has some good insights, even if the genre is of debatable taste. A few brought chuckles to me. But, I feel it is a bit like laughing at Yes, Minister.



Friday, January 22, 2016

Spin-orbit coupling and "triplet" superconductivity

My collaborators and I just finished a paper

Spin-orbit coupling and odd-parity superconductivity in the quasi-one-dimensional compound Li0.9Mo6O17
Christian Platt, Weejee Cho, Ross H. McKenzie, Ronny Thomale, and Sri Raghu

Here is the abstract.

We welcome any comments.
One thing I learnt and found interesting what the unusual spin-orbit coupling that arises due to lack of inversion symmetry in the four-atom unit cell. I will post separately about that next week as the story of the corresponding coupling in graphene is an interesting one.

Thursday, January 21, 2016

What should my Ph.D advisor expect from me?


The relationship between a Ph.D student and their advisor or supervisor is a complex one. I have written before about the importance of both parties having clear expectations and communicating and agreeing on them at the beginning of working together.

Over the years I have seen or heard of some strange things. Below is a list of some things that students have found their advisor expected.

work on my thesis topic

prepare a slide or two about my thesis work that they will present at a conference

write a small grant application for me to fund my travel to a conference

give feedback on their draft grant application that is related to my project

give one of their lectures while they are away at a conference

organise a weekly seminar series

act as de-facto supervisor for an undergraduate student research project that is not related to my project

maintain the group website

spend a morning on the registration desk for a conference they are organising

referee a paper or grant for them

perform maintenance duties around the lab

ignore an off-colour joke they make on just one occasion

maintain the group computing facilities

help paint the lab

"protect" them at all times

teach two weeks of their class while they are away

add "honorary" co-authors to a paper I wrote

provide data, calculations, or text for a paper which I will not be a co-author

prepare slides for a talk they will give that largely does not involve my thesis work

maximise citations to their papers, even when not relevant

promote their pet theories, even when I disagree

spend a whole week editing the abstract book for a conference they are organising

accept the job offer that they want me to take

not contradict them in public

submit a paper they have not read

write a whole grant application for them [see the second last paragraph here].

paint their house

come into the lab on a sunday morning

mow their lawn

babysit their children [Linus Pauling did!]

buy dinner for one of their visitors

listen to long monologues about their political or religious views or their personal life or departmental politics

not report unethical behaviour

let them confide deep personal struggles and emotional issues

let them post 86 poems about me on their Tumblr page

let them verbally abuse me

let them sexually harass me

have sex with them

The list starts off in a reasonable way and then steadily moves towards the unethical.
It is a slippery slope. Somewhere it moves from being a "team player" to working in a sweatshop, i.e. exploitation and abuse of power.

Where do you draw the line? Not everyone will agree on how far down the list the line should be drawn. I would probably draw it after the fourth item. Some of the lower items I might ask a student to do (sometimes with financial compensation), but not expect them to do it. Others will go a bit lower, perhaps particularly experimentalists.
I welcome comments.

How should students deal with this issue?

First, it is important to research an advisor before you sign up. Ask their current and former students. If there is "weirdness" involved find someone else. Don't think it will be any better for you. Second, if they seem o.k. agree on things before you start work.

Finally, if you are in an unacceptable situation get help. Don't endure it.

How should advisors deal with the issue?
See where you are on the list. Initiate discussions with your students.

Wednesday, January 20, 2016

The Sommerfeld model is a Pauling point

A basic question that comes up in introductory solid state physics is:
Why does the Sommerfeld model for metals work so well?
It assumes that electrons are non-interacting fermions. Yet if you calculate the first order correction (in e^2 where e is the electronic charge) in the Coulomb energy you find it is comparable to the kinetic energy associated with the ground state.

Aside: the success of Sommerfeld is such a puzzle that Wigner mentioned it (for the wrong reasons in my view) at the end of his famous 1962 essay, The Unreasonable Effectiveness of Mathematics in the Physical Sciences.

The standard answer we give students is screening  plus Landau's Fermi liquid theory.
However, an interesting question is what happens if you try to actually do some sort of systematic many-body expansion with respect to the Coulomb interaction. Can you get the calculation to converge to experiment and see why Sommerfeld is good?

In Telluride last (northern) summer I heard a nice talk by Timothy Berkelbach that is relevant to this issue. The message I took away was that the Sommerfeld model is a Pauling point, i.e. by accident it gets the right answer for the wrong reasons.

The relevant paper has now appeared on the arXiv.

Spectral Functions of the Uniform Electron Gas via Coupled-Cluster Theory and Comparison to the GW and Related Approximations 
James McClain, Johannes Lischner, Thomas Watson, Devin A. Matthews, Enrico Ronca, Steven G. Louie, Timothy C. Berkelbach, Garnet Kin-Lic Chan

A key graph is below. It shows the quasi-particle dispersion relation for a uniform electron gas with r_s=4, the value relevant to sodium.

For comparison the "binding energy" [i.e. the k=0 energy] deduced from experiment is -2.6 eV, and the values for Sommerfeld and LDA are about -3.1 eV. Hartree-Fock gives -7.3 eV!

As you increase the "level of theory" [i.e. the sophistication of treatment] of electron correlations you go from Sommerfeld to HF to HF+GW to CCSD [Coupled Cluster Singles and Doubles].
Then one sees the answer at first gets worse and then improves and you almost get back to where you started!

Aside: This also illustrates how LDA is a Pauling point too!

Tuesday, January 19, 2016

Who am I representing on this committee?

I think this is an issue that is rarely discussed.
People just assume their point of view is the valid one. Then they are surprised or upset when the "representative" does not act as they "should".

Suppose Professor Smith from the Chemistry department is on a committee from the College of Arts and Sciences that determines some internal funding or policies for the departments within the College. Here is a list of different interests and perspectives that she could represent:

-her own
-her research subfield of synthetic organic chemistry
-the chemistry department
-just the science departments
-all the departments in the College
-the Dean of the College
-senior management of the university
-tax payers, alumni, students., ..

Idealists might say that of course she should consider and understand all views. But that rarely happens. Some seem to think that in this cruel world representatives should fight tool and nail for their own interests, or at least those of their department. This may mean belittling others.

My own view is that representatives each bring a unique perspective; but, they should be striving to do what is in the best interests of the institution as a whole. Unfortunately, this sometimes means going against the interests and views of their department and of senior management.

What do you think?


Monday, January 18, 2016

Infrared spectroscopy: What is the Condon approximation?

How do you calculate the absorption intensity associated with a molecular vibration?
First, why might you care?
This is not just a basic scientific issue that is only of interest to people working in molecular spectroscopy.
It actually lies at the heart of global warming. For example, why is methane a much worse greenhouse gas than carbon dioxide? It is because it has a much larger infrared (IR) absorption intensity in the relevant frequency range.

In the electronic ground state consider a transition from a vibrational level with quantum number j to one with i. The absorption intensity is given by


where the dipole matrix element between the two vibrational states is
I  use r to denote all the nuclear co-ordinates.
mu_g (r)  is the dipole moment of the molecule in the electronic ground state.
For notational simplicity I neglect the vector character of the dipole moment.

One can now make an approximation to greatly simplify evaluation of this matrix element and to provide some physical insight. One approximates the dipole moment by its first derivative term in a Taylor expansion.
This is known as the Condon approximation.

Aside: I can't find the original reference. Please let know if you know it.

This is a very useful approximation. First, it give some insight.

a.
It tells us that the IR intensity is dominated by the variation in the dipole moment of the electronic ground state with nuclear co-ordinates.

b.
If the nuclear wave functions are harmonic, then the only non-zero IR transition is that of the fundamental (i.e. from the ground state i=0 to the first vibrational excited state, i=1). There are no overtones, i.e. higher harmonics. This is known as the double harmonic approximation. (The first is the Condon approximation).
In reality, all potential energy surfaces are slightly anharmonic and so this leads to the presence of weak overtones in IR spectra. Their intensity can be used to estimate the amount of anharmonicity, both in the potential and the dipole moment surface (i.e. deviations from Condon).

Second, the Condon approximation makes calculations of intensities a lot easier. One does not need to calculate the full dipole surface, mu_g(r) just its first derivative at the equilibrium geometry. This is what almost all computational quantum chemistry codes do.

How reliable is the Condon approximation?
It seems to be very good for most molecules. Corrections are often only a few percent.
Here is one study by Juana Vazquez and John Stanton.
One can measure vibrational frequencies extremely accurately (especially in the gas phase), e.g. to within 0.01 per cent. In contrast, one can usually only measure vibrational intensities to within about 10 per cent. This provides less motivation to worry about corrections to Condon.

However, there are exceptions. Jim Skinner and collaborators have shown how for the OH stretch in liquid water one needs to take into account the dependence of the dipole moment on the nuclear co-ordinates of the surrounding water molecules.

Thursday, January 14, 2016

Stunning and creative microscope images where science meets fashion

The New York Times has a nice obituary Michael W. Davidson, a Success in Microscopes and Neckwear, Dies at 65

I did not know Davidson personally but I did benefit from his art. I visited the National High Magnetic Field Lab at Florida state several times in the 1990s. In appreciation my host Jim Brooks gave me a few of the neckties [which I still wear, on the rare event I actually wear a tie!] and a series of prints of images of  Australian products such as that of Vegemite below. I still have these prints on display in my office.


The website Molecular expressions contains not just a gallery of many beautiful microscope images but also more technical discussions about microscopy.

One thing I did not know about Davidson that I learnt from the obituary was the important role he played in the work for which the Chemistry Nobel Prize of 2014 was awarded.

Friday, January 8, 2016

Is private for-profit education always a disaster?

John Quiggin lists a litany of cases where the answer to the question is yes. It particularly makes for depressing reading given the blind push for privatisation by the Australian government.

I think for-profit private education is particularly likely to be problematic when two criteria are met:

1. Consumers are not paying up front. For example, if they are paying tuition from government loans that they will only have to pay back a decade later.

2. Consumers (students and parents) don't have the ability and/or information necessary to assess the quality of what they are getting for their money. For example, a poor parent who never finished high school may struggle to know what a good ollege education should look like.

These criteria means the "free" market is not very responsive.

Until a month ago I thought that the answer to this question was universally, Yes!
However, one should never say never.

For my birthday my son gave me a copy of The Beautiful Tree: A Personal Journey Into how the World's Poorest People are Educating Themselves by James Tooley.
It is a fascinating story of how Tooley discovered how in many slums in the Majority World there were private for-profit schools, typically charging of the order of $1 per week (roughly 10% of parents income, seriously!) for tuition. Many development "experts" from the West and local government education administrators denied the existence of these schools. When finally presented with evidence  of their existence they debated their effectiveness and moral value. However, Tooley showed the quality of the education they present is higher than local "free" government schools. This is because the latter are remarkably ineffective due to government corruption, unconditional tenure of teachers, teacher absenteeism, .... Uneducated poor parents then make well informed decisions to sent their children to the private schools which are very accountable to parents because they are dependent on fee income. Remarkably most of these schools waive or reduce tuition for about 20 per cent of their students who are deemed to be particularly economically needy!

My son became aware of the book through this podcast on Econtalk.
Some of the issues were dealt with in a cover story on The Economist.