[Mitch]

I'm preparing a story for chemistry-blog.com regarding this Langmuir paper and was wondering if anyone else has seen it, read it, or would like to comment on it. Since it makes a rather bold claim, I would like some feedback.

Link to paper:
http://pubs.acs.org/doi/abs/10.1021/la802430k

[renge ishyo]

The paper was not bad overall. It was interesting to see that the the rate of neutralization in water in these experiments was so slow. I am still curious to see how the rate would be affected by completely removing any and all electrodes from solution (I didn't see that they did this, all the experiments at least kept some electrodes in solution), and watching how quickly the dye neutralizes on it's own. The overly long delay could very well be caused by slowly changing residual buildups of charges near or around the electrodes placed in solution.

However, the author's claim at the way end that this experiment demonstrated the violation of electroneutrality was absurd. The principle of electroneutrality states that if you consider the liquid as a whole that the net charge within the liquid will be neutral (http://www.answers.com/topic/electroneutrality-principle). The principle does not state that if you subdivide a liquid and only consider a portion of its volume that the charge in that portion has to be neutral. That's how the authors interpreted the principle, and their interpretation of the principle is what is wrong as far as I can tell.

[Borek]

1. Speration of pH - H⁺/OH^- transport is diffusion controlled. No idea how fast the diffusion is, but pH gradient during electrolysis seems to be the obvious effect of limited diffusion speed. Could be it is not - but it should be calculated to check.

2. pH gradient implies they have concentration cell. Same problem as obove - before explaining potential difference by "charge sperataion" they should prove observed results are not effects of concentration differences.

However, the author's claim at the way end that this experiment demonstrated the violation of electroneutrality was absurd. The principle of electroneutrality states that if you consider the liquid as a whole that the net charge within the liquid will be neutral (http://www.answers.com/topic/electroneutrality-principle). The principle does not state that if you subdivide a liquid and only consider a portion of its volume that the charge in that portion has to be neutral. That's how the authors interpreted the principle, and their interpretation of the principle is what is wrong as far as I can tell.

Electroneutrality principle states that volumes large enough are neutral - and my take is that "large enough" means femto- or attoliters if not even smaller volumes. If they are able to prove points 1,2 posted above are not sufficient to explain observations, we can revisit electroneutrality.

[renge ishyo]

The electroneutrality principle is just the conservation of electrical charge from Physics in disguise. Just as you can "violate" the conservation of charge by isolating only one small part of a system you can "violate" electroneutrality by considering only one small part of a liquid. As an extreme example, let's say that I define a volume to be no bigger than a single H₃0⁺ molecule. If I can define any volume I please then I can do that, which means I've just violated electroneutrality! My volume only contains a positive charge! Nobel prize here I come!

Except I don't think the definition of electroneutrality works that way...if you consider that the liquid in the paper had charge separation then yeah one side of the liquid is not electrically neutral because it is positive and the other side is negative. However, I would bet that the liquid as a whole is still neutral; i mean if you were to individually count the charges on all the molecules in the positive region of the liquid and all the molecules in the negative region I would bet that the net charge of the liquid as a whole is neutral. This is what the principle says. If it didn't say that it would be a meaningless principle that would be violated every time you dumped chunks of salt into a liquid (there would be small localized charge fluctuations in various regions of the liquid until equilibirum was established in that case as well).

[Borek]

I don't understand your post - as it seems to me most of what you have wrote is not in any way in contradiction with what I have wrote - with the exception of one small point, which is irrelevant to the electroneutrality definition.

Before we divide the solution into single molecules, at which moment electroneutrality will be violated for obvious reasons, we can divide it into blocks of - say 10¹⁸, 10¹², 10⁶, 10³ molecules. The real question that arises is - when does the electroneutrality starts to be violated? Obviously it depends on the concetration of dissolved substances. As I wrote earlier, my bet is that (regardless of the solution composition) volumes in the femtoliters range are neutral (that means volume containing tens of millions of molecules). Probably even much smaller volumes are still electroneutral, although statistical fluctuations are going to play more and more important role. And it follows that solution as a whole, being sum of smaller eletcroneutral blocks, is electroneutral as well.

if you consider that the liquid in the paper had charge separation

I don't have to at the moment, I have proposed two obvious explanations to the observed phenomena.

(To be honest I am almost sure that on the level of single electron charges no solution is electroneutral - but that's another story).

[Mitch]

My write-up on the paper is here:

http://www.chemistry-blog.com/2009/09/03/electroneutrality-is-dead/

[renge ishyo]

I don't understand your post - as it seems to me most of what you have wrote is not in any way in contradiction with what I have wrote - with the exception of one small point, which is irrelevant to the electroneutrality definition.

I wasn't debating your point Borek; in fact i was using your point about a minimum volume to highlight an absurdity in that if you can define your own subvolume for electroneutrality then you can always arrange things to violate the principle anytime you please. The question of beyond what minimum volume can electroneutrality not be violated even within a larger liquid is an interesting one, and warrants its own experiments. Even if some pattern was found in those experiments it still wouldn't change the fact that electroneutrality itself still holds since by definition it is stated for liquids as a whole (I keep bolding it to emphasize this point!). You *can't* violate it by subdividing a liquid, because the principle says nothing about what will happen when you do that.

I read your article Mitch, you were even harder on them than me  A few things to mention, the setup with the two beakers separated was a control only and was not the basis of their discussion as far as I can tell...they were talking about charge separation within a single large tub where the positive and negative ions were connected in the same liquid (Figure 1).  Also, when they removed the charging electrodes they seemed to always make sure that "other electrodes" were already in solution before the removal took place (or at least I couldn't find an instant where this wasn't the case). Your description made it sound like they discovered more than they did!

Their challenge of electroneutrality seemed to imply that if we took a clean beaker and scooped out some of their positively charged liquid from one side of the tub, and used a separate beaker to scoop out negatively charged liquid from the other side of the tub, that we could then connect the two beakers with a wire and  get current to flow *with no salt bridge present*. This is implied because if the liquid in the left beaker really did have an excess of + charge, and the liquid in the right beaker had a net excess of - charge then there would be no need for counter ions to flow through the salt bridge to get the electrons to flow from the negative beaker to the positive one.  The reason the salt bridge is needed to begin with is to maintain electroneutrality in each liquid; you would need the ions to flow through the salt bridge because the net charge in each liquid is always neutral. If the liquids were not always neutral this would not need to be the case. Therefore, showing that you can get electric current to move from one beaker to the other without a salt bridge is the necessary requirement to show that that the liquids indeed have overall charges which are not neutral. I did not see them perform this experiment to back up their hypothesis. My prediction is that such an experiment wouldn't back up their hypothesis, but they can go ahead and try it and let me know how it goes...

Still, it seems like they made this particular claim not in the name of scientific rigor but in order to possibly garner fame and attention for their paper. So mission accomplished. I just think it was a mistake, it undermined the more interesting aspect of their paper which is that you could maintain charge separation within a large body of water for an extended period of time.

[Borek]

Even if some pattern was found in those experiments it still wouldn't change the fact that electroneutrality itself still holds since by definition it is stated for liquids as a whole (I keep bolding it to emphasize this point!). You *can't* violate it by subdividing a liquid, because the principle says nothing about what will happen when you do that.

OK, now that it is obvious we agree when it comes to the general idea I will explain why you are wrong

Imagine we have started with the solution - volume V - of water.

This solution contains V*10^-7 moles of H⁺ and exactly the same amount of OH^- (that's not true, but it will became obvious why later).

Now, we divide the solution pouring exactly half of it to the other vessel.

Most of the solution was divided equally, and most of the volume was already neutral, but every ion that was close to the border through which the solution was divided, was either left in the original vessel, or has been poured to the new one with probablity 0.5. p=0.5 - probability that H⁺ is in the first vessel, q=0.5 - probability that H⁺ is in the second vessel. p+q=1. And you know what? This is Bernoulli distribution. That means that - while statstically ions have been divided exactly in half - in reality they were not, as the variance in this case is something like pq. Thus both solutions are charged. Charges we are talking about are pretty small, as number of ions involved was pretty small when compared with the total number of ions in the solution, but nonetheless they exist and their presence is an inevitable result of the fact that world is in fact granular, not continuous.

Now it should be obvious why the original solution has been already charged - it was prepared by splitting other solution.

[renge ishyo]

But could you prove that assertion experimentally Borek? If the number of the ions are so small, then it is true that nobody can really say what happens at that level because we can't measure it. However, since we can't test it it is easy to come up with another proposition that is just as reasonable but claims the exact opposite outcome.

For instance, let us say that we did the same experiment and tried to divide a solution so that one solution has an excess of positive charge over the other. The problem is what will happen at the *interface* where the single solution eventually breaks apart into two. Can it break off so that one of the two solutions has even a little teeny tiny excess of charge over the other? It could very well be that if one of the solutions had a net charge over the other at some region where the liquid was breaking apart that these charges could quickly flow back across this interface in such a way that the break cannot occur unless the two sides were exactly electrically balanced.

That is a reasonable proposition, because the movement of the charges is not completely random as they would be in a true statistical distribution...regions of the water where there is temporary distinct charge separation would tend to come together and resist separation into two liquid parts with much more urgency (the electrical force is strong!) than neutral regions of the liquid where a break could occur between the two parts with the charges balanced on each side equally. Therefore the break of the solutions into separate beakers would be much more likely occur with the charge exactly balanced such that the two regions in question were neutral before and after the break. This explanation has the added advantage in that it doesn't violate the macroscopic experiments that electroneutrality was based on. It also sucks because it too is impossible to test 

[Borek]

Perhaps some version of Millikan experiment measuring electron charge - but with the use of water drops? Oil that Millikan used in the original version have not contained ionized substance, charge was just electrostatic, effect of rubbing. In the case of water - if there is a charge separation - one may expect different statistic of the charges; most likely higher average charge of a single drop. And the average charge should depend on the concentration of salts dissolved in the water.

[renge ishyo]

I really dunno; it would be a very challenging experiment that is for sure and a difficult hypothesis to test.

At any rate I don't think the people who wrote the paper put this much thought into how difficult it would be to actually disprove electroneutrality. They seemed to give the idea of the presence of negative counter ions in the low pH region of their liquid about as much consideration as they gave to the nature of the "mysterious gases" that were emerging at the two electrodes.

[Mitch]

renge ishyo: I over simplified the design of their experiment, because their actual design is just plain stupid. Since they did use it as the control, and since I didn't discuss their results, and since it is an easier figure to explain what they attempted to do, I decided to just go with it.

[renge ishyo]

renge ishyo: I over simplified the design of their experiment, because their actual design is just plain stupid. Since they did use it as the control, and since I didn't discuss their results, and since it is an easier figure to explain what they attempted to do, I decided to just go with it.

Fair enough