A thesis should have a thesis

Yes. I heard a humanities scholar speak about this today. Indeed, according to Wikipedia, "the word "thesis" comes from the Greek θέσις, meaning "position", and refers to an intellectual proposition."

So, what is the main thesis of your Ph.D? Are you ready to defend this thesis?

The challenge of energy sustainability

Today at the I-CAMP School in Brisbane, Dan Cox, Director of I2CAM gave a nice talk about climate change and future energy options.

Dan has taught an undergraduate course on such issues for the past 13 years at UC Davis. (Unfortunately, the course web page seems to be down right now).

He mentioned some really nice lectures on the same topic by Nate Lewis (Caltech)

A few things I found interesting.

Graphs of energy consumption vs. GDP tend to plateaus at high incomes. i.e, affluent countries can increase their "standard of living" without increasing their energy consumption. Japan and EU use about half the energy per capita of USA/Canada, yet have comparable standards of living.

Electricity consumption is 4 times greater in Wyoming than in California.

Oil is used almost 100% for transportation. Alternative electricity sources will not reduce dependence on Middle East oil without electric cars.

M. King Hubbert was a petroleum engineer who predicted world oil production would peak in 1970. The production rate is related to the derivative of the solution to the logistic differential equation, which describes many phenomena that initially increase exponentially until finite resources limit their growth.

[Aside there was actually a West Wing episode, The Hubbert Peak]

Can you do this with your tap water?

CO2 dissolving in ocean decreases pH and destroys coral. Even if global temperatures do not increase due to increased CO2, the increase in acidity of the oceans will wreak environmental havoc.

Searching for better solar cell dyes

This morning I looked over a nice review article on Metal free organic dyes for dye-sensitized solar cells. Here are a few highlights and comments.

The figure below is a schematic of what one is trying to design: a dye molecule D-pi-A in which upon excitation by a photon the excited state (D-pi-A)* decays non-radiatively into a charge separated state D+pi-D- which leads to charge injection into the TiO2.

Many of the dyes have the structural motif of methine dyes and are prone to cis-trans isomerisation in their excited state. It is implied that this is a bad thing one wants to stop. I actually thing an optimised/directed isomerisation could be good because it can lead to a twisted intramolecular charge transfer (TICT) state.

I am curious to what extent many of these dyes can be described by the type of effective Hamiltonians that Seth Olsen and I (and also independently by Anna Painelli and collaborators) have been exploring for other methine dyes. This could help answer questions about TICT states and design principles.

The following summary point by the authors is noteworthy:

It is therefore important to determine design rules for organic dyes to achieve improved properties and performance so that they may in the future compete or supersede ruthenium(II) sensitizers. Clearly, the development and optimization of materials for organic solar cells in general is not (yet) rational, but rather empirical. This is due to the multiple parameters which have to be taken into account when novel dyes and materials are designed for organic solar cells. The efficiency of the solar cells does not only depend on the molecular structure of the dye, but sometimes even more so on solid-state properties, such as aggregation, morphology, and self-assembly. In the case of DSSCs, the efficiency additionally depends on the type of photoelectrode, anchoring group of the dye, electrolyte, and mediating redox couple.

To me it raises questions about whether

-so much effort should be expended on making new dyes rather than on understanding solid state interactions.

-authors should be more cautious about attributing variations in device efficiency to properties of specific dyes rather than solid state properties.

Theoretical chemistry in the Rockies

Apparently chemists got jealous of the Aspen Center for Physics and so started the Telluride Science Research Center.

I am looking forward to going there in three weeks for a workshop on Condensed Phase Dynamics. It will be great because I will get to meet some theoretical chemists whose papers I have read but not met yet, including Garnet Chan, Phillip Geiser, Dimitrii Makarov, Bill Miller, Peter Rossky, Greg Voth, .....

Quantifying the effect of "small" chemical changes

Physicists say the details don't matter. Chemists say they do. Biologists say the details are a matter of life and death.

If you make a "small" change in a molecule what effect will it have on its properties?

Long ago Hammett found a fairly robust and empirical way [the Hammett equation] to quantify the answer, at least for organic chemical reactions involving aromatic molecules.

A nice J. Phys. Chem. A paper by Cordes et al. considers the following important problem in photophysics. Consider the molecule below which upon irradiation can undergo a conformational change (photoisomerisation).

If the substituents R1 and R2 are changed what effect does that have on the rate of photoisomerisation (non-radiative decay)?

They find the rate of both the forward and backward photochemical reactions [which varies by two orders of magnitude] can be correlated with Hammett's parameters for the substituents.

Fermions or bosons?

In a "round table discussion" about the theory of the cuprate superconductors Patrick Lee suggested that the genuinely new idea that has been developed is:

"the notion of emergence of gauge fields and fractionalized particles as low-energy phenomena in systems that did not contain them in the starting model."

He suggests that this idea is of comparable importance in condensed matter theory to that of Goldstone bosons.

Gauge fields emerge when the electron or spin operators are represented in an alternative manner such as in terms of Schwinger bosons, slave fermions, slave bosons, or slave rotors. But a key question is for a given model Hamiltonian, which is the appropriate representation.

For quantum spin models it seems that which side of the Charles River you work on determines your preference for a particular representation. At Harvard, Subir Sachdev favours bosonic spinons, while on the opposite of the river, at MIT Patrick Lee favours fermionic spinons.

Where do the gauge fields come from?

Exam post mortem

This week I marked the exams for the undergraduate course I helped teach this past semester. I usually find this a depressing exercise. It painfully reveals how little I have managed to teach students and how little they have learnt. Note, I consider the responsibility is joint.

The last three questions in the exam are from the condensed matter part of the course I teach. The students have seen the questions before, either in tutorial problems, assignments, or past exam papers. Yet they struggle to do them.

Some of the questions are meant to test not just knowledge but also general skills such as converting units, summarising results in a figure, and sketching curves. But, this is what I find particularly disappointing. Students still have trouble doing some of these things, even at second year university.

I think next time I teach the course I particularly need to keep hammering home that all physical quantities have units, and that you need to keep track of them in calculations. Furthermore, I want the tutors to take off marks on assignments when students don't properly keep track of units.