Thanks for the mention above. I've made a silver ion electrode, which used in conjunction with a commercially available reference electrode, is giving me numbers that are off quite a bit from the calculated values. Well, it's just a first trial, and I expect that things will get better with some more experimentation.
But it looks like there will be a working solution for those that are interested in pursuing it.
Ok, here is a summary of several months work and a list of equipment needed.
First, you need a high impedence VOM such as the Radio Shack 22-812, which reads in the MV range or +/- 1V on the total scale and which can reverse polarity automatically or via a switch so that you can span the range from -300 to +300 mv. as quickly as possible.
Second, you need a silver electrode of about 20 grams weight and quite thick (about $15 - 20 depending on the spot price of silver). This would be good, heavy quality silver wire or rod of 99.99 fine. Yes, that is 99.99 fine. Anything less will cause problems at some time or another.
Third, you need a double junction Silver, Silver Chloride, KCl electrode such as the HI5132 or HI73127. You will need filling solution for the former, and storing solution for both.
That should allow you to set up your experiments quite nicely. It will not help you make emulsions unless you build a bridge and I am working on that right now.
Now, with average cleanliness, I get results as follows:
With 0.1 M NaBr I have a range of -36 mv to -70 mv with lots of noise. With scrupulous cleanliness and fresh electrodes, conditioned and cleaned with my best effort, I get -65 to -69 mv, with the calculated value of -62.6 mv.
With 0.1 M NaCl I have a range of +22 mv to +50 mv with lots of noise. Cleaning the equiment brings my range down to 33 mv to 36 mv, with the calculated value of 36.5 mv.
The moral of this is that cleanliness is essential. The Silver electrode corrodes and the reference electrode does not like distilled water very much and must be kept conditioned. The type of reference electrode is not important, nor is the bridge (at this time - still working on the bridge though - more later).
Since the reference electrode leaks saturated KCl, this is not a suitable method for making an emulsion. The bridge is critical for best results.
I hope to summarize the 1 M salt results and the KI results in the near future for you.
At the present time, it seems that you will need 3 Silver electrodes, one each for the 3 major salts. Also, you will need 3 bridges for the same 3 major salts. And, this will not work very well in the presence of any silver halide solvents such as Ammonium Ion or the like. This is more due to the uncertainty of calculation of a correct value than anything else.
BTW, I have verified these results with 4 VOMs, with and without stirring, but all at 20 deg C. And, in the final circuit, to stabilze the system, I may have a battery and potentiometer to allow for zeroing the system with a standard salt along with a slope control for changing salt concentrations from high to low. At the present time, if I zero the system on a standard salt, I often see a 20 mv offset (slope) as the salt becomes more and more dilute. I am working to figure out if this is a slope problem or a contamination problem or both.
PS, the above, when detailed, has added about 6 pages to the book. You see? I have not forgotten it either.
Que es VOM?
Also, when you say a bridge, is this a chemical or an electronic function?
Have you ever looked into the Vernier brand of measurement probes? They are designed for classroom use, and carry a "don't use these for real science" warning, but they are very easy to use, work with a lot of different platforms (win/mac, TI calculators, some old Palms), and have several components for adapting any BNC-type commercial probe. They are very "plug&play" and would be a logical choice for the home user, IF they are accurate enough for your purposes. They also have a very large installed base, lots of High School science teacher fans, and the company does a lot of work creating supporting curriculum, projects, activities and training seminars. You can also hook them up to Lego Mindstorms robot kits :cool:! There are also software & hardware development kits. They are also reasonably priced. Jolly good fun.
One of the folks who advertises of APUG who sells a compensating developing timer (I think) uses the Vernier usb-enabled temperature probe in his design.
VOM = Volt-Ohm Meter, a term to describe these meters.
The equipment I am using is known to work with the measurement I am trying to take. Just any old probe will not do. Sorry.
A bridge is a salt or wire connection between two solutions, each of which contain a solution to be measured. A post was made here earlier describing the entire procedure. Each end of the bridge is called a junction.
Here is a summary of the KI tests.
By inserting the electrodes and going ahead with the measurement, my range is about -222 to -313 mv for a 0.06 M solution. The calculated value is -276.5 mv
My range with super clean, scrubbed and conditioned electrodes is -265 to -282 mv. These are all at -20 deg C. I can do better! I am working on an easy method of cleaning and conditioning the electrodes.
Now back to the NaBr tests....
My results over 2 days for 0.1 M NaBr now ranges from -62 to -63.2 mv, calc -62.6. This is pretty good right now. The 1.0 M NaBr results over 2 days is -116.2 to -120, calc -116.9. So, I am pretty close to where I want to be in measurement methods and lab technique (work flow).
The bridges are made and working, I have worked out a gel and salt solution, I have worked out a method of filling them, and a method of storing them. There is still a lot to do.
Is the noise you observe random by nature and does the value change quickly? If so, the value could be averaged using analog (RC circuit) or digital (if the system is later connected to PC) system to obtain even more stable data.
Well, actually, I can damp out the noise in 2 ways. One is by using an analog meter. Then the inertia of the needle damps out digital fluctuations (another benefit of analog). The needle simply does not move fast enough. Another way is to ballast the circuit by using 2 1.5 V AA batteries and a pot and center the reading on zero. In the mv range, and with such low current, the system is hypersensitive. Waving the probes in the air can give me readings varying from -100 to +100 mv sometimes, but it varies with the phase of the moon. ;)
At EK we used some sort of custom amplifier and lots of grounds to prevent this type of noise. At home, I am just improvising. Unfortunately, I never knew the contents of the black boxes we used.
Another part of the noise is the conditioning of the Silver electrode and the cleanliness. We "bought" our electrodes prepared for us, and then conditioned them with standard salts and used another black box. Again, I am improvising as I go along.
I should mention that there are 2 types of noise, if you read into the above what it implies. There is a rapid fluctuation back and forth and a slow drift up or down. The former is due to the low current and voltage, and the latter is due to conditioning and deconditioning of the electrode.
There is another separate problem, and that is with my Potassium Nitrate. It is not pure enough and tests positive for a tiny level of KCl in it. This affects the action of the bridged system during long use. It is used very dilute though, which reduces or minimizes the effect of the impurity on the system during measurement.
However, purchase of pure KNO3 is restricted in the US at the present time. So, I have to use what I can get right now.
Well, the noise is gone as are most of the other problems. It was the conditioning that was not sufficient.
Here are the correct conditions for 12 guage Silver wire. They are stronger for the 1 oz billet.
4 minutes treatment in 1 M NaBr with 3.5 V from two AA batteries. The Silver electrode you are going to use is connected to the positive (red) pole of the batteries and the negative pole (black) goes to the donor electrode which is also Silver of the same purity. The Silver electrode will blacken significantly during this 4 minute conditioning / plating. The electrode must be then washed with distilled water and must be stored under distilled water. It will last for at least 24 hours. After that, the readings begin to become unstable again.
From the start of the conditioning, the operation and storage must be under reduced light as you are creating a light sensitive surface on the Silver electrode.
Insure that all sides of the electrode are plated evenly and slowly over 4 minutes. If you plate more than about 12cm of #12 wire, you may have to use longer times or higher voltages. I find that a 20 gram billet requires 9V over about 3 - 4 minutes.
If your readings are low, you have overdone the conditioning, and if they are high, you have underdone the plating. You must have an even coating on the Silver electrode which does not flake off during use. This was one of my problems.
My results in two tries? 1: -115.5 and 2: -116.1, Calculated is -116.96. These were at 20 deg C using 1.0 M NaBr with stirring.
Cleanliness and proper storage along with the right conditioning seems to have fixed up all problems. My thanks to Kirk Keyes for 6 inches of 99.99 fine #12 Silver wire.
Well, this is more good progress!
I think the conditioning and keeping the "sensitived" electrode in the dark is a great tip that one may not ever figure out on their own...
You say "reduced light" - do you mean safelights or are dim lights good enough?
Any reason not to use a stainless or carbon rod for the "donor" electrode/cathode (negative terminal of the battery)?
If anyone is interested where I found the 12 gauge silver wire, it was here:
http://www.ccsilver.com/silver/superfines.html#four - follow the link for the ".9999 wire"
I used subdued light and it seemed to work just fine. This is a two bulb system in my lab as opposed to the 8 bulbs I normally use, so it is a lot less light.
I wanted to go down stepwise.
As for the other electrode, I am not sure. I used silver to keep contamination at a minimum. All I can say is "give it a try". I will be trying other materials myself. I would not recommend "poison" metals though such as copper, zinc, lead and tin, and I would worry about stainless, as Iron contamination will fog emulsions as well. Some carbon has adsorbed sulfur compounds that are not nice either. It depends on the source of the carbon. So, testing is obviously in order. Since plating is taking place, I try to avoid anything that might be carried over in the stream to plate out.
So, the bottom line is "use what works". I know Silver works.