I've conducted a simple experiment on a glass sample that has two clean copper electrodes biased with 20v DC. Between the two electrodes I have placed a drop of de-ionised water with Universal Indicator added.
Over a period of time the electrodes changes colour, as expected, with the cathode going purple (OH-) and the anode going red (H+).
After a longer period of time the colours start to spread out toward the opposing electrodes, pH gradient. However, the amount the purple (OH-)spreads is apx three times greater that that of the red (H+).
I have repeated this 10 times in random positions and I get the same results.
I would have thought that the H+ ions would have more mobility than the OH- ions and thus I would have expected the opposite to happen.
Can anyone help and through some light on this phenomena, or have I missed something?
You shouldn't interpret the data in terms of H+ OH- concentrations without taking into account how the color of the indicator changes with changes in each of those ions. And that is a nearly impossible task without quantitative measurements.
Regarding mobilities, keep in mind that electrical neutrality must be present even loaccly. If H+ is moving, something else must also move to maintain local electrical neutrality.
The indicator surely has a neutral and an ionized form. The neutral form diffuses at one rate. The ionized form diffuses faster under the influence of the electric field.
To test this, take some acid function ( -COOH ) and some amine function (-NH2 ) indicators and see if they behave in opposite ways.
I'm away from my references or I'd give some examples.
Raw hydrogen ions cannot exist naked in solution. They combine with a nearby hydroxyl molecule to form H3O. You are actually generating copper ions at the anode, not hydrogen ions. The process is
At the anode, a copper atom gives up two electrons to become an ion:
Cu(s) - 2e --> Cu(++)
At the cathode, water dissociates and hydrogen ions accept electrons to form hydrogen gas which escapes:
2H2O --> 2H(+) + 2OH(-)
2H(+) + 2e --> H2(g)
So for every copper ion, two hydroxyl ions are produced.
This is why telephone circuits use negative polarity. If the voltage was positive, small leakage currents would cause the wire to disappear through electrolysis.
You can calculate the amount of copper ions in solution by knowing the current and the time interval. The calculations are highly prone to error due to the numerous unit conversions needed, but a dos program written by Roger Schafley called Mercury will do the conversions for you. Here is an example calculation I use for a much larger electrolysis cell:
; Colloidal Copper Generator ; Calculations Bob Lee's method ; Note change ; k = 0.5* 63.5 / 96485 ; Coulombs required per gram of copper
; Roger Schafly's Mercury is available at ;
formatting link
;
formatting link
;
formatting link
Cou = I * sec ; total number of Coulombs esec = I / 1.60217733e-19; electrons per second gm = k * I * sec ; Faraday's equation isin = esec / sqin ; ions per sq. in. per sec isnm = isin / 6.45e14 ; ions per square nanometer per sec k = 0.5* 63.5 / 96485 ; Coulombs required per gram of copper lt = 3.785 * gal ; convert gallons to litres lt = ml / 1000 ; convert millilitres to litres mg = gm * 1000 ; convert grams to milligrams ml = 29.57 * oz ; convert ounce to milliliters phr = ppm / hrs ; ppm per hour ppm = mg / lt ; 1 ppm is 1 milligram per litre sec = hrs * 3600 + mnt * 60 ; convert hours to seconds uAin = 1e6 * I / sqin ; current density in uA per sq in
hrs = 1 I = 3.111e-3 ; current 02 3.3k 3.111 mA ml = 1450 ; volume of dw mnt = 0 ; minutes sqin = 9.5 ; wetted area
In this example, 0.00368 grams of copper were released giving a concentration of 2.54166 ppm (parts per million). It turns out the maximum you can achieve is about 3ppm before the copper starts plating out on the cathode.
As far as the change in color, I posted the results of two experiments using silver ions that show how to make these ions visible:
CS> Making Ions Visible
formatting link
Re: CS> Making Ions Visible
formatting link
In my example, the silver ions seemed to have the same mobility as the hydroxyl ions, since the color bands appeared to meet in the middle in the first experiment.
Since the mobility is a function of the size of the ion and the applied voltage, I would expect the copper to move much slower than the hydroxyl. This appears to be the reverse of what you observe.
Perhaps someone at sci.chem can offer an explanation.
Of course, I should say a nearby water molecule. From Chemtutor:
"In a way, there is no such thing as a hydrogen ion or proton without anything else. They just don't exist naked like that in water solution. Remember that water is a very polar material. There is a strong partial negative charge on the side of the oxygen atom and a strong partial negative charge on the hydrogen side. Any loose hydrogen ion, having a positive charge, would quickly find itself near one of the oxygens of a water molecule. At close range from the charge attraction, the hydrogen ion would find a pair (its choice of two pairs) of unshared electrons around the oxygen that would be capable of filling the its outer shell. Each hydrogen ion unites with a water molecule to produce a hydronium ion, H3O+, the real species that acts as acid. The hydroxide ion in solution does not combine with a water molecule in any similar fashion. As we write reactions of acids and bases, it is usually most convenient to ignore the hydronium ion in favor of writing just a hydrogen ion."
Thanks for your detail response. I have re-run the test using graphite electrode and 18.3MG pure water with Universal Indicator. I get the same results. From what you are saying I guess the H+ ion and the OH- ions are attaching themselves to something in the Universal Indicator?
The reason is probably because H+ *does* move faster than OH-. H+ has a limiting ionic conductivity of about 350 S.cm^2/eq and OH- has a limiting ionic conductivity of about 200 S.cm^2/eq. Limiting ionic conductivities are directly proportional to the mobility (consult any physical chemistry text). They are 4-7x more mobile than any other ion due to the Grotthuss transport mechanism. Glass provides plenty of -OH groups, and even more importantly, loads of adsorbed water (which I doubt you controlled for), to allow this mechanism to occur.
Oops I misread..OH- more than H+, so this has nothing to do with ion mobility. Also, there is a lot more going on here than H+ and OH- migration if you add universal indicator, which usually contains a brew of various indicators that are quite large molecules. I think you need to consider the mechanism of reaction between the relevant indicator(s) in acid/base. Just for example, methyl orange is a large diimide contains a sulfonic acid group that changes from red to yellow on deprotonation. Red/purple sounds like two different indicators to me - you could easily have two different charged large indicator molecules moving towards each electrode.
Yes, graphite definitely eliminates the copper electrolysis problem:)
As you point out, and another poster mentions, the effect may be due to the indicator. It might be nice to try the experiment in a shot glass to get a larger volume and work in 3D instead of a thin film.
Another thing you might consider is using plain red cabbage juice. If you get it right, you can end up with wisps of indicator scattered through the solution, with pure dw in between. The wisps of indicator will light up as the various ion species cross them. The hydroxyl ion causes the solution to turn purple, but I don't know what H3O(+) would do. This should allow you to estimate the diffusion rate in pure water with less interference from possible reactions with the indicator.
You can use the Faraday calculations mentioned in my earlier post to calculate the number of ions in solution. If so, you need to use a constant current source (perhaps 100uA to 10mA) with sufficient voltage compliance to prevent saturating, and change the Faraday constant from copper to whatever ion you are interested in tracking.
Adding to all other replies: universal indicator response is not necessarilly linear, so I doubt you are really able to tell what pH and pOH are at which solution place.
Best, Borek
--
http://www.chembuddy.com - chemical calculators for labs and education
BATE - program for pH calculations
CASC - Concentration and Solution Calculator
pH lectures - guide to hand pH calculation with examples
I check sci.chem from time to time. They have a few good people, but if they are not active there's little else worth reading. Pretty much the same as here on s.e.d.
What prompts you to investigate the mobility of hydroxyl and H3O+ ions? This can turn into very interesting research. Water seems so simple, but it has to be one of the most complex substances on the planet. For example, you probably have seen "The anomalous properties of water", at
formatting link
There are a few movies on the web showing electrolysis of water. Here's one using phenolphthalein as an indicator. It's pretty crude, but it might give you something to compare with your results. The url is
I am working on trying to understand the mechanics of AC (alternating current) growths on water contaminated printed circuit boards. So the first stage is to understand the mechanism under DC the move onto AC. Although, I have produced carbonate type residues and dendrites (at low freq) on circuit boards using AC. The idea is to use the graphite and Universal indicator to help shine some light on what is happening, more than the cancelling out theory. E.g. the ion moves from the electrode on the positive going cycle and on the negative going cycle it returns and cancels out. No very true as the ion will prob. never go back to the same place it left.... I have a history of question on sci.chem.electrochem:
From: "WAYNEL" - Find messages by this author Date: 10 Jul 2005 09:25:26 -0700 Local: Sun,Jul 10 2005 5:25 pm Subject: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Remove | Report Abuse
I've conducted a simple experiment on a glass sample that has two clean
copper electrodes biased with 20v DC. Between the two electrodes I have placed a drop of de-ionised water with Universal Indicator added.
Over a period of time the electrodes changes colour, as expected, with the cathode going purple (OH-) and the anode going red (H+).
After a longer period of time the colours start to spread out toward the opposing electrodes, pH gradient. However, the amount the purple (OH-)spreads is apx three times greater that that of the red (H+).
I have repeated this 10 times in random positions and I get the same results.
I would have thought that the H+ ions would have more mobility than the
OH- ions and thus I would have expected the opposite to happen.
Can anyone help and through some light on this phenomena, or have I missed something?
From: Marvin - Find messages by this author Date: Sun, 10 Jul 2005 15:00:27 -0400 Local: Sun,Jul 10 2005 8:00 pm Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Report Abuse
- Hide quoted text -
- Show quoted text -
WAYNEL wrote:
You shouldn't interpret the data in terms of H+ OH- concentrations without taking into account how the color of the indicator changes with changes in each of those ions. And that is a nearly impossible task without quantitative measurements.
Regarding mobilities, keep in mind that electrical neutrality must be present even loaccly. If H+ is moving, something else must also move to maintain local electrical neutrality.
Reply
Aubrey McIntosh, Ph.D. Jul 10, 10:49 pm show options
From: "Aubrey McIntosh, Ph.D." - Find messages by this author Date: Sun, 10 Jul 2005 16:49:41 -0500 Local: Sun,Jul 10 2005 10:49 pm Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Report Abuse
WAYNEL wrote:
The indicator surely has a neutral and an ionized form. The neutral form diffuses at one rate. The ionized form diffuses faster under the influence of the electric field.
To test this, take some acid function ( -COOH ) and some amine function
(-NH2 ) indicators and see if they behave in opposite ways.
I'm away from my references or I'd give some examples.
From: Mike Monett - Find messages by this author Date: Sun, 10 Jul 2005 22:42:06 -0400 Local: Mon,Jul 11 2005 3:42 am Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Report Abuse
- Hide quoted text -
- Show quoted text -
Wayne wrote: > I've conducted a simple experiment on a glass sample that has two > clean copper electrodes biased with 20v DC. Between the two > electrodes I have placed a drop of de-ionised water with Universal > Indicator added. > Over a period of time the electrodes changes colour, as expected, > with the cathode going purple (OH-) and the anode going red (H+). > After a longer period of time the colours start to spread out > toward the opposing electrodes, pH gradient. However, the amount > the purple (OH-) spreads is apx three times greater that that of > the red (H+). > I have repeated this 10 times in random positions and I get the > same results. > I would have thought that the H+ ions would have more mobility > than the OH- ions and thus I would have expected the opposite to > happen. > Can anyone help and through some light on this phenomena, or have > I missed something? > Cheers > WayneL
Hi Wayne,
Raw hydrogen ions cannot exist naked in solution. They combine with a nearby hydroxyl molecule to form H3O. You are actually generating copper ions at the anode, not hydrogen ions. The process is
At the anode, a copper atom gives up two electrons to become an ion:
Cu(s) - 2e --> Cu(++)
At the cathode, water dissociates and hydrogen ions accept electrons to form hydrogen gas which escapes:
2H2O --> 2H(+) + 2OH(-)
2H(+) + 2e --> H2(g)
So for every copper ion, two hydroxyl ions are produced.
This is why telephone circuits use negative polarity. If the voltage was positive, small leakage currents would cause the wire to disappear through electrolysis.
You can calculate the amount of copper ions in solution by knowing the current and the time interval. The calculations are highly prone to error due to the numerous unit conversions needed, but a dos program written by Roger Schafley called Mercury will do the conversions for you. Here is an example calculation I use for a much larger electrolysis cell:
; Colloidal Copper Generator ; Calculations Bob Lee's method ; Note change ; k =3D 0.5* 63.5 / 96485 ; Coulombs required per gram of copper
; Roger Schafly's Mercury is available at ;
formatting link
;
formatting link
;
formatting link
Cou =3D I * sec ; total number of Coulombs esec =3D I / 1.60217733e-19; electrons per second gm =3D k * I * sec ; Faraday's equation isin =3D esec / sqin ; ions per sq. in. per sec isnm =3D isin / 6.45e14 ; ions per square nanometer per sec k =3D 0.5* 63.5 / 96485 ; Coulombs required per gram of copper lt =3D 3.785 * gal ; convert gallons to litres lt =3D ml / 1000 ; convert millilitres to litres mg =3D gm * 1000 ; convert grams to milligrams ml =3D 29.57 * oz ; convert ounce to milliliters phr =3D ppm / hrs ; ppm per hour ppm =3D mg / lt ; 1 ppm is 1 milligram per litre sec =3D hrs * 3600 + mnt * 60 ; convert hours to seconds uAin =3D 1e6 * I / sqin ; current density in uA per sq in
hrs =3D 1 I =3D 3.111e-3 ; current 02 3.3k 3.111 mA ml =3D 1450 ; volume of dw mnt =3D 0 ; minutes sqin =3D 9.5 ; wetted area
In this example, 0.00368 grams of copper were released giving a concentration of 2.54166 ppm (parts per million). It turns out the maximum you can achieve is about 3ppm before the copper starts plating out on the cathode.
As far as the change in color, I posted the results of two experiments using silver ions that show how to make these ions visible:
CS> Making Ions Visible
formatting link
Re: CS> Making Ions Visible
formatting link
In my example, the silver ions seemed to have the same mobility as the hydroxyl ions, since the color bands appeared to meet in the middle in the first experiment.
Since the mobility is a function of the size of the ion and the applied voltage, I would expect the copper to move much slower than the hydroxyl. This appears to be the reverse of what you observe.
Perhaps someone at sci.chem can offer an explanation.
From: Mike Monett - Find messages by this author Date: Sun, 10 Jul 2005 23:27:21 -0400 Local: Mon,Jul 11 2005 4:27 am Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Report Abuse
Mike M> Hi Wayne,
Of course, I should say a nearby water molecule. From Chemtutor:
"In a way, there is no such thing as a hydrogen ion or proton without anything else. They just don't exist naked like that in water solution. Remember that water is a very polar material. There is a strong partial negative charge on the side of the oxygen atom and a strong partial negative charge on the hydrogen side. Any loose hydrogen ion, having a positive charge, would quickly find itself near one of the oxygens of a water molecule. At close range from the charge attraction, the hydrogen ion would find a pair (its choice of two pairs) of unshared electrons around the oxygen that would be capable of filling the its outer shell. Each hydrogen ion unites with a water molecule to produce a hydronium ion, H3O+, the real species that acts as acid. The hydroxide ion in solution does not combine with a water molecule in any similar fashion. As we write reactions of acids and bases, it is usually most convenient to ignore the hydronium ion in favor of writing just a hydrogen ion."
From: "WAYNEL" - Find messages by this author Date: 11 Jul 2005 03:10:42 -0700 Local: Mon,Jul 11 2005 11:10 am Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Remove | Report Abuse
Hi Mike
Thanks for your detail response. I have re-run the test using graphite
electrode and 18.3MG pure water with Universal Indicator. I get the same results. From what you are saying I guess the H+ ion and the OH- ions are attaching themselves to something in the Universal Indicator?
From: Mike Monett - Find messages by this author Date: Mon, 11 Jul 2005 11:08:11 -0400 Local: Mon,Jul 11 2005 4:08 pm Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Report Abuse
WAYNEL wrote:
Hi Wayne,
Yes, graphite definitely eliminates the copper electrolysis problem:)
As you point out, and another poster mentions, the effect may be due to
the indicator. It might be nice to try the experiment in a shot glass to get a larger volume and work in 3D instead of a thin film.
Another thing you might consider is using plain red cabbage juice. If you get it right, you can end up with wisps of indicator scattered through the solution, with pure dw in between. The wisps of indicator will light up as the various ion species cross them. The hydroxyl ion causes the solution to turn purple, but I don't know what H3O(+) would do. This should allow you to estimate the diffusion rate in pure water with less
interference from possible reactions with the indicator.
You can use the Faraday calculations mentioned in my earlier post to calculate the number of ions in solution. If so, you need to use a constant current source (perhaps 100uA to 10mA) with sufficient voltage
compliance to prevent saturating, and change the Faraday constant from copper to whatever ion you are interested in tracking.
From: Mike Monett - Find messages by this author Date: Tue, 12 Jul 2005 05:27:23 -0400 Local: Tues,Jul 12 2005 10:27 am Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Report Abuse
WAYNEL wrote:
Hi Wayne,
s=2Ee.d.
I check sci.chem from time to time. They have a few good people, but if
they are not active there's little else worth reading. Pretty much the same as here on s.e.d.
What prompts you to investigate the mobility of hydroxyl and H3O+ ions?
This can turn into very interesting research. Water seems so simple, but it has to be one of the most complex substances on the planet. For example, you probably have seen "The anomalous properties of water", at
formatting link
There are a few movies on the web showing electrolysis of water. Here's
one using phenolphthalein as an indicator. It's pretty crude, but it might give you something to compare with your results. The url is
formatting link
(560k)
It is from Step 3 at:
formatting link
Mike Monett
Reply
WAYNEL Jul 12, 2:53 pm show options
Newsgroups: sci.chem, sci.chem.electrochem, sci.chem.electrochem.battery, sci.electronics.design From: "WAYNEL" - Find messages by this author Date: 12 Jul 2005 06:53:25 -0700 Local: Tues,Jul 12 2005 2:53 pm Subject: Re: OH- spread more than H+ ions? Reply | Reply to Author | Forward | Print | Individual Message | Show original | Remove | Report Abuse
Hi Mike
I am working on trying to understand the mechanics of AC (alternating current) growths on water contaminated printed circuit boards. So the first stage is to understand the mechanism under DC the move onto AC. Although, I have produced carbonate type residues and dendrites (at low freq) on circuit boards using AC. The idea is to use the graphite and Universal indicator to help shine some light on what is happening, more than the cancelling out theory. E=2Eg. the ion moves from the electrode on the positive going cycle and on the negative going cycle it returns and cancels out. No very true as the ion will prob. never go back to the same place it left.... I have a history of question on sci.chem.electrochem:
My mission is to dig deeper into the subject. You are correct conformal coatings et al can prevent it but not necessarily eliminate it. There are certain conditions that can be avoided that can prevent certain forms of migration over a specified product life such as track thick/distance and not using metals such as silver.
then stop either getting board or their bosses wanting an immediate answer. I have years so.... The story will continue.
Another indicator that might be interesting is disodium fluorescein. This is the stuff that makes antifreeze the dayglow green color. It is safe enough to inject into blood or pour into rivers for tracing.
It changes from an extreemly efficient fluorescent dye to colorless and non-fluorescent when it gets protonated.
Wouldn't it be better to seal the board with any of several overcoats, then package it so water can't reach it? If you are in a humid environment, you might also have to deal with various acids caused by pollution. These would cause severe corrosion on pcb conductors, and probably would damage sensitive components mounted on the board.
formatting link
I looked at some of the posts. As mentioned above, it would probably be a good idea to protect the board and prevent moisture from reaching it.
As far as your observation of copper dendrite growth with DC, this is a fairly well-known phenomenon. In a conventional electrolysis cell, the copper ions migrate to the cathode. When the ion concentration begins to exceed about 3 ppm, the ions start plating out on the cathode. I have watched this growth many times.
The tendrils seeme to grow out from the cathode until they reach the anode. The spacing between anode and cathode was about 1 inch in my experiments. The electrodes were about 3 or 4 inches long, and multiple tendrils would grow from the cathode.
The tendrils are conductive, so the first one to reach the anode would short out the cell and stop the electrolysis.
I never had occasion to use AC, but you can easily visualize the ions leaving the anode, then returning to plate out when the electrode becomes the cathode. In this fashion, dendrites could grow from both electrodes and eventually cause a short.
On a regular pcb, as well as copper ions, you would have tin, nickel, and probably other metals from component leads. Any acids in the water would only accelerate the corrosion.
I don't believe there is much you can do to stop this process except coat the board with suitable protection and package it so moisture can't reach it.
Interesting. You mean I can go out to my car, drain some antifreeze, and get a ph indicator? That would be nice.
I'm not sure what you mean by the word protenate in this context. Does it mean the indicator responds only to hydronium ions, or would any postive ion such as metal work?
Any other suggestions for cheap sensitive indicators, preferably ones you can find in a grocery store or pharmacy on a weekend?
Not all antifreeze use this same dye, but in principle, yes. As an aside, when ethylene glycol (of antifreeze) oxidizes it can make oxalic acid. Oxalic acid and iron can proceed to make "greensalt" which is almost the same color as antifreeze with the disodium fluorescein.
Most pH indictors are discussed in the context of either H+ or Na+ (OH- in water, with Na+ as a spectator ion) If metals that are not in the first two columns are used, you results have a lot more nuance than the standard language provides. For example, when I cook cabbage in an aluminum container, the color is off, and the container shows signs that some aluminum oxide is removed. I assume that the Al+3 is being complexed by the dye in the cabbage, but this is anecdotal -- I haven't collected data.
One easy dye to obtain is the broth from cooking red cabbage. It is red when acidic, blue when basic, and lavender in betwixt.
Bromthymol blue is another easy to obtain pH indicator. It is sold in aquarium shops as a treatment for "ick."
Litmus paper itself uses the juice from lichens. It sticks in my mind that the Scientific American "Connections" column attributes Litmus paper to a McIntosh in Scotland.
Beet juice is mentioned at
formatting link
but I have not tried it. I do know that when beets are cut into 1/4" cubes and cooked while protected from air in a pressure cooker, the juice is a dayglow red. I speculate that this is pH sensitive, I just haven't had the desire to add lye or battery acid to my beets.
Other foods: cranberry, elderberry, tumeric, grape, and blueberry are mentioned at
Wayne, I'm sure there is plenty in that topic to keep you busy for a long time:)
Another consideration is whiskers. Primarily tin, but can be other metals. This is discussed from time to time in sed. Here's one url:
formatting link
BTW, I tried to use your email address to see if you have a home page somewhere. Did you know your url points to a nonexistent Yahoo group, then goes to a german isp, and finally ends up in England? Here's the links:
<A HREF="
formatting link
">
formatting link
formatting link
formatting link
That yahoo link is amusing. It is totally unexpected, and would definitely stop most people:)
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.