I’m trying to get hold of

Roeland van Wijk, Saskia Bosman, Eduard P.A. van Wijk. The Journal of Alternative and Complementary Medicine. 2006, 12(5): 437-443. doi:10.1089/acm.2006.12.437.

- which apparently reproduces some of Rey’s results.

My Univ doesn’t subscribe… not that it’s a sudden anti-woo conversion, more that we don’t seem to take Liebert journals.

“Is it a safe assumption that you can infer anything about the structure of a liquid from studying the frozen liquid?”

You might be on reasonable grounds to infer something about the structure of a liquid crystal from its solid state, crystal structure, in combination with other techniques. For an isotropic liquid like water, though, you can only really make inferences about the sort of structures you might observe at interfaces between that liquid and another phase of matter, be that gas, solid or another, immiscible liquid. These will be pretty dynamic structures in any event, even if the molecules involved exchange positions with others at a rather lower rate than occurs in the bulk.

So much for the woo-friendly premise of the paper. Criticisms of the science would be a lack of controls, a failure to assign the peaks to the relaxation of particular excited states and a disturbingly arbitrary approach to radiation intensity selection.

My explanations? Woolly speculation – perhaps I should steer clear – but…

Using heavy water implies they are looking at a heavy isotope effect, so coupling to O-D stretching vibrations is probably important. Because water has basically two stretching modes, of differing symmetries (called symmetric and asymmetric stretches), I suspect we are seeing two populations of electronic excited states, of corresponding symmetries to the vibrations. For luminescent decay to be an “allowed” quantum transition, symmetries must coincide (or be related through a consistent mathematical operation – I only know a bit of photophysics…) As the temperature rises, in whichever sequence, first one of the vibration modes is thermally possible, so the corresponding excited state can decay through that state with a corresponding photon emission. The second vibration becomes allowed at a higher T, whereupon the second excitation can decay by luminescence.

What the author observes, therefore, are variations of emission from the ice, and it is very interesting that the lower T event is pretty much constant whatever the solute. The inferences to be drawn from the author’s data are (a) increasing colloid content increases this higher T emission with a roughly logarithmic effect, (b) that dissolved ionic solids cause an decrease of this emission with increasing concentration.

This is where the really interesting question is posed, and it would be great if someone was in a position to try and answer it. Why do ionic solutes quench, whilst the colloid intensifies the thermoluminescence at around 160K? I bet it’s down to direct coordination of O atoms to cations (and H atoms to chloride) in the ionic solutions, whilst water will build a relatively robust hydrogen bonded structure (yes, I know!) around colloid particles. Goodness knows how, though…

http://en.wikipedia.org/wiki/Thermoluminescence_dating

- but it seems to be a technique where proper physicists have to build the equipment themselves. And application to frozen liquids seems rarer than application to solid samples, or to sediments.

Any physicists on the forum care to offer a helpful general comment on thermoluminescence?

What I have also been trying to work out is which of Rey’s papers describes the low-temperature thermoluminescence technique and (critically) the equipment used to make the measurement in full detail. It is not the one about the homeopathic dilutions, which simply cites earlier his own earlier papers. Gamma or electron beam irradiation of the frozen substance (e/g/ water) prior to controlled warming seems to be a critical part of the basic measurement technique.

Is it a safe assumption that you can infer anything about the structure of a liquid from studying the frozen liquid?

Were the results reproducible? The author says someone else has reproduced them, but what I really want to know is what is the variability of the samples. If I test 100 different samples of, say, 30C Nat. Mur., how much would the spectra vary? There’s no information on this in the paper.

The claim seems to be that the graphs look different for different homeopathic ‘remedies’ [Figure 3]. However, the peaks appear at roughly the same temperatures, although they have different amplitudes. If you were seeing different structures, would you not expect to see emission at different temperatures, rather than just variation in amplitude?

As I say, not my field.