jump to navigation

Photocatalysis: Buy my snake oil September 30, 2009

Posted by calvinus in Energy, Green Chemistry, Photocatalysis.
Tags: ,

Was at a one-day meeting the other day.  There were couple of companies trying to sell themselves and their products. Photocatalysis was mentioned a couple of times in despatches.

Photocatalysis, for those that do not know, is a way of using light (usually sunlight) to drive a “useful” chemical reaction. In the case below, this could be splitting water into oxygen and hydrogen, but it might also involve turning organic chemicals into water and carbon dioxide.

I should say that it is telling that I see this figure appear quite often but I do not know who was the originator of the image.

Normally the photocatalyst is a semiconductor and the vast, vast majority of scientific articles published use nanoparticles of titanium dioxide to do this.  The popularity of titanium dioxide is partly historical (Fujishima and Honda discovered the phenomenon using titanium dioxide[1]) but mostly due to titanium dioxide’s efficiency at converting light that has enough energy to a chemical product.

And therein lies an important drawback: the light must have enough energy to overcome barriers in the semiconductor.  For titanium dioxide, this means the light has to be particularly high in energy – it can only use ultraviolet light.  Given that this is only around 4% of natural sunlight, this is a problem if photocatalysis is to be a useful process.  This is especially the case for indoor applications where room lights tend to have negligible ultraviolet light.

Photocatalysis Book

There are, it would appear, companies out there that claim to be able to do everything through photocatalysis.  They can:

  • destroy nasty pollutants in water
  • make glass, concrete, roads, buildings that clean themselves
  • kill bacteria such as “E. coli, MRSA, swine flu, bird flu”
  • destoy spores and turn them into carbon dioxide (!)

Some of the claims are fanciful, to say the least, but there are some basis in these claims.  Destruction of organic pollutants has been known for a while and substantially researched.  Self-cleaning glasses are known (see St Pancras station for an example of their use).  Anti-bacterial propoerties are again reported in the scientific literature although the jury is out – I don’t believe anyone fully understands the possible mechanism and I have certainly NEVER seen any reports of photocatalysis being used to combat the bacteria du jour, swine flu or bird flu.  You might, if you were being charitable, sugest that these claims may have been the result of someone not fully understanding the background science.


However, what triggers my abject cynicism is that one company went as far to say that their photocatalysts worked in both outside applications AND indoors.  When asked how the performance compared, indoor and out, I was very surprised to hear them openly admit that they hadn’t actually tested it.  So how could they make these claims?  “Er…”

This narks me.  This narks me a lot. Not only does it border on fraudulent, it does give the whole area an air of “buy my snake oil”.  For those who are trying to do good, honest work in this area, it pollutes the pool of potential investors and cheapens what could be a useful area of renewables research.

Of course, preying on people’s fears (“Kills MRSA and Swine Flu dead”) is one thing, but would they use the same bombast if they were a pharmaceutical company and had people’s lives hanging on their results?

ResearchBlogging.org [1] Fujishima A, & Honda K (1972). Electrochemical photolysis of water at a semiconductor electrode. Nature, 238 (5358), 37-8 PMID: 12635268


French blogs, pt II September 27, 2009

Posted by calvinus in Blogs.
add a comment

Just found one (whilst looking for something else).

Parlez-vous Chimie.

Can’t be the only one, can it?

Software & a lack of understanding thereof September 20, 2009

Posted by calvinus in Uncategorized.
Tags: ,
add a comment

I have a theory.  Granted, this theory has only had a comparatively short period of evidence gathering to find supporting information, but it is a theory nonetheless.

When I was going through the education system (have I ever really left), the mantra was that we were much more computer literate than all of the professoriate and lecturing staff.  Anywhere else I have been, the mantra has been the same: the students are more IT savvy than the staff.  Even now I have colleagues that learn from our students.  This is worrying given how clueless many of our students are in matters of IT.

Let me be quite clear.  It is not their fault, nor is it a criticism of our students’ intellects or abilities.  Yes “da yoof” are more connected, more technologically fluid (but not fluent) than ever before.  However, faced with a minor problem that isn’t on the menu bar of a MS program, “da yoof” are lost.  I get exactly the same questions from panicking and utterly frustrated students that I get from my mother-in-law (and given that she still uses a dial-up, this might give you an idea of the level).

So, here is the theory.  Looking back, I suspect computer literacy probably peaked with the generation that was overly familiar with Windows 2000/ME (not, I stress, that I am suggesting that these were the ideal OS!).  After around that time, operating systems started becoming more user-friendly inasmuch that they would do a lot more things automatically and intuitively for the “average” user (NDLR: the lowest common denominator).  Couple this with the ubiquity of PCs, and MS in particular, and general understanding of the structure and function of software dropped.  The theory goes that understanding and ability to use software (or anything) is inversely proportional to the slickness and popularity of the software and operating system (which means that bloodyVista is a ray of hope for the improvement of user understanding!).

Some examples of what I mean.

1) When did the “My documents” function arrive?  Every year I teach a basic statistics course which involves the heavy use of computer-based workshops.  Despite the fact that the students love the statistics, every week I encounter the same problems.  Student X says that they have a problem – they cannot find the file that they saved their work in last week.

JC “Where did you save it?”

SX “Dunno. It was in My Documents last week.”

JC “Which My Documents?”

SX “?”

JC “My documents on the hard drive or My Documents on the network drive?”

SX “The what?”

JC “OK, have you tried searching for the file?”

Suffice to say, the answer is usually “no” as they hadn’t even realised that the “find” or “search” function existed… So, if something doesn’t automatically appear when you want it, it is clearly a problem of the computer, or so it would appear many people think.

2) Post-graduate student Y tries new software for operating a bit of kit.  Said software saves everything automatically by default and in a proprietary binary that I couldn’t care less about.  This is a good thing as you have all the instrument parameters, all the data, all the conditions, etc. in one compact file.  But this isn’t good enough for student Y: you can’t open up the data in Excel.  This is considered to be a “bad thing” and all hope is lost and there is darkness upon the face of the earth.  “This is useless, I can’t open my data in Excel.” I decided that a response of “RTFM” may not be the most constructive and after a small amount of fiddling (i.e., three clicks) we were able to get the data as ASCII that will happily open up in any spreadsheet that our IT department permit us to use (go figure the length of that list).  But this requires a conscious decision and effort on the part of the operator, even if it is just three clicks.  Therefore it was suggested that the software was “hardly the most user friendly”.

Lets pause for a minute.  Are we regressing?  Should all new software only save data in ASCII human-readable text?  Shall we take this to its logical conclusion and ask that our software only accepts input files that are text-only?  No GUIs please?  This is ironic given that I know the carnage that causes – I teach the use of computational chemistry programs (although admittedly I should use something other than vi).  While we are at it, should we not go back to UNIX and DOS??

The problems, I can handle.  I am more worried about the inability, or lack of confidence, to fiddle around and find possible solutions.  I suspect this year I shall mostly be teaching from XKCD: