Oxidation-Reduction Potential (ORP) is a net measure of the electrochemical potential of the water, given the constraints of the platinum electrode that is used to measure it. It is a thermodynamic measurement so gives an indication of what is possible, not how quickly it will or can occur. This is a critically important distinction. Thermodynamics predicts that oxygen will oxidize most organic material (including your skin), but fortunately this occurs very slowly unless the temperature is raised where above a critical temperature the reaction produces enough heat to sustain itself as with burning/fire. The limiting factor that distinguishes what WILL happen vs. would COULD happen is the reaction rate and this is determined in each direction by the activation energy barrier of the reaction (and of course, the temperature and concentration, really activity, of reactants). Though chemicals with higher ORP do tend to oxidize more chemicals even at room temperature, it is not an absolute rule. So all statements claiming that high ORP means anything in terms of general oxidation or disinfection are simply incorrect. Whether oxidation or disinfection actually occurs depends specifically on the chemical substances themselves.
Chlorine is a fairly selective oxidizer and mostly reacts with nitrogenous organics and ammonia (more generally, mainly amines, reduced sulfur moieties or activated aromatic systems). Potassium monopersulfate (MPS) does not react quickly with ammonia, but reacts with carbon-carbon double-bonds to produce epoxides (oxygen bridging the two carbons), reacts with some amino acids and registers strongly as ORP though it is not nearly as strong a disinfectant as chlorine at pool temperatures.
In addition to higher temperature, another way that reaction rates can be increased is by lowering the activation energy of the reaction as occurs with catalysts including enzymes. This is the essence of many chemical reactions in the body where the energy potential found in the combination of oxygen with complex organics can be converted into controlled and stored energy in the form of phosphate bonds (as with ATP) that can then be used to drive other chemical reactions. In short, oxygen + food ---> stored energy ---> new chemicals.
ORP for many standard half-reaction chemicals follows the theoretical Nernst equation which predicts that a 2-electron transfer reaction will have an increase of 8.9 mV for every doubling of concentration of the reactant getting reduced or an increase of 29.6 mV for every 10x increase in concentration. The chlorine half-reaction is a bizarre exception since it has an increase of around 28 mV per doubling in chlorine concentration or 93 mV per 10x increase in chlorine concentration. This implies a 0.64 electron transfer which, of course, makes no sense.
As described in this post, the ORP mV per doubling of chlorine concentration varies significantly by sensor/manufacturer and by water source. Also see this post where two sensors measuring the same water varied by 100 mV or more in 23% of the pools that had more than one device. Even Clifford White's respected Handbook of Chlorination shows a huge variance. I could find no source explaining these discrepencies (especially the variance against the Nernst equation mV per doubling). Suffice is to say that chlorine chemistry is complex and that ORP should not be relied upon as an absolute standard except in the broadest rough terms when chlorine is involved.
So how can ORP be reliably used? The answer is for process control. One can create a setpoint for an ORP level corresponding to a Free Chlorine (FC) level in a particular pool. ORP does correlate most strongly with the hypochlorous acid concentration which is the primary disinfectant in the bulk pool water, so ORP is more reliable than looking at FC alone, but if one knows the Cyanuric Acid (CYA) concentration and sets the FC relative to the CYA level, then the ORP reading at the desired level can be used as a setpoint for process control. An amperometric sensor or chlorine-specific sensor may be used to validate the FC/CYA relationship since it is possible for some other organic chemicals to behave like CYA and bind chlorine somewhat loosely having it register as FC instead of CC.
There are reports that direct sunlight especially near the sensor can result in a lower ORP level. This is ironic since the active chlorine level is not significantly depleted and in fact when chlorine breaks down by the UV in sunlight it produces hydroxyl radicals that are even more powerful oxidizers (so should register as higher, not lower, ORP), though are very short-lived. Hydrogen gas from saltwater chlorine generators likewise results in a lower ORP reading. So for process-control, one should not place the ORP sensor in-line with the output of a chlorine generator or too close to returns in the pool.
The link didn't work. Please repost. I plan to do the ORP write-up this weekend if I get the chance.
Sorry, I was trying to do this on my I-pad. Sometimes it works, and sometimes it doesn’t
That is an excellent document and one that people should read first for detailed info on ORP. Unfortunately, there are a few mistakes in the document that make it confusing or misleading so let me clear those up, but overall it really is an excellent treatment on ORP.
The initial discussion correctly notes that an oxidizing agent itself get reduced (and vice versa). However, in describing the redox reactions in Table 1, he wrote:
A negative value for a reaction means that the reaction is "energetically favored," i.e. it wants to go in the direction of left to right.
which is exactly the opposite of what it should be. The values in the table are "reduction potentials" which mean that a larger (more positive) value is more likely to get reduced itself so is a stronger oxidizer of other chemicals (i.e. it can oxidize other chemicals). The most negative reactions in the table tend to go from right to left; the most positive from left to right. Of course, any can go either way depending on what it is paired against.
Another problem is that he shows the Nernst equation using a natural logarithm, but then refers to RT/zF with z=1 and T=298K as being 59.16 mV, but that is what you get when using a base-10 logarithm so ln(10)*RT/zF; he left out the ln(10)=2.303 factor in his derivation (the 59.16 mV is correct and for a 2-electron reaction as with chlorine, this is 29.58 mV for every factor of 10 in concentration difference).
Another problem is that he wrote that the voltage of the ORP sensor will increase when "we turn up the heat on our sample", but the opposite is true from looking at the Nernst equation since one subtracts the term that has RT/nF so higher T lowers ORP. He says that oxidation speeds up at higher temperatures, but ORP is not measuring reaction rates, but a thermodynamic quantity. Also, the Eo base value is temperature dependent and can go either way depending on whether a reaction is exothermic or endothermic.
The major flaw in his paper, however, is implying that ORP sensors are similar across manufacturers. He explains how there can be differences in reference electrodes, but the manufacturer's own tables of ORP vs. chlorine concentration not only show differences in absolute values, but in slope as well. He implies that the slope should be the same based on the Nernst equation, but not only is the slope different, but it's way, way higher than what the Nernst equation predicts. Note that the 59.16 mV is for a 10x concentration difference and a 1 electron transfer. The chlorine reaction is two electrons and for a doubling of concentration the Nernst equation predicts 8.90 mV. As shown in this post, the slopes vary all over the place. Also, as shown in this post, even two sensors from different manufacturers measuring the same pool water can vary by 100 mV or more, not just the much smaller 23 mV he shows.
Also, calculating theoretical ORP from the Nernst equation results in far higher numbers even after subtracting out the 230 mV for the silver/silver chloride reference electrode. A hypochlorous acid concentration of 0.05 ppm at 77°F and pH 7.5 with 350 ppm chloride has a theoretical ORP relative to silver/silver chloride of 922 mV while registers on an Oakton ORP sensor (from the link I gave) of 658 mV. There's no way that there is that the difference is explained by reducing substances. Basically, neither the absolute ORP nor the slope match theory at all being 28 mV vs. 9 mV. I have never seen an explanation from anyone about this, yet ORP is promoted as if it is some magical measuring device. It's fine for process control and a rough order-of-magnitude guide of water quality, but that's about it.
OK, I've finished the main discussion post. Let me know if you have any questions or have additional information you want in the post.