CR-Scientific: Minerals, "Home Lab" & Earth Science Newsletter: October / November 2003 issue

CR-Scientific Minerals & Experimental Science newsletter

Terms of Use

Issue #7
HTML version

In this issue:

I. Electrolysis Experiment
II. Becquerels to Sieverts?
III. An Interesting Trick With Equilibrium
IV. Site News

I. Electrolysis experiment

Restorers of antique vehicles sometimes use a process they call "electro rust removal". As one would gather from the name, it involves electrolysis and is intended for iron or steel items. It seems well-suited to hollow objects such as fuel tanks (thoroughly emptied of the gasoline and all its fumes) where there's no other effective way of getting at the rust.

The basic process uses sodium carbonate solution as the electrolyte. Into this solution goes a stainless steel rod to which the positive electrode of a battery charger is connected. Update 1: because some hexavalent chromium (Cr⁶⁺) will form at the stainless steel anode, it is better to use an iron or a carbon steel anode instead of stainless. Nearly any anode containing chromium will produce at least some Cr⁶⁺. Unless you are trying to make Cr⁶⁺ on purpose, it's probably better to avoid forming it.

The negative electrode of the charger is connected to the tank and care is taken never to let the steel rod (+) electrode touch the tank directly. No more than 4 to 8 amperes at 12 volts are allowed to flow through the circuit; a car battery is not recommended because of the lack of current limiting.

The chemistry of the process seemed interesting. It was instinctively apparent that reduction of ferric iron (Fe⁺³) must be taking place, but what else was going on in this process? How did it work?

As with many "simple" experiments, it turned out to be much more complex than one might guess. For an an overview of what happens chemically, have a look at "The Chemistry of Cleaning Rusted Iron by Electrolysis" by Bill Tindall and Spencer Hochstetler, available at http://www.holzwerken.de/museum/links/electrolysis_explanation.phtml. For a more hands-on but less technical treatment of the procedure, see "Bill's Electrolysis Page" by Bill Dickerson at http://antique-engines.com/electrol.asp. There is a photograph of the technique in action.

What would happen if, instead of a rusted cathode and a steel rod anode, both electrodes were platinum and didn't participate directly in the reaction-- as would be the case in the Brownlee electrolysis apparatus. Of course, all we'd be doing then would be performing electrolysis of a sodium carbonate solution, but this deceptively simple-looking process might be very interesting in itself. What gases would be produced? Would there be hydrogen and oxygen alone, or oxygen and carbon dioxide, or all three? Would there be others? Why / why not? What would happen to the pH of the solution? What effect would increasing or decreasing the sodium carbonate concentration have on the products? (The answers to these are not intuitive. There are multiple variables-- some not obvious-- which can affect the outcome.)

Some Electrochemistry: It is important to point out that "electro rust removal" is thought to work primarily by the action of gas bubbles. Hydrogen forms at the cathode (2H⁺ + 2e^- ----> H₂), which in our case is a rusted piece of iron or steel. The tiny bubbles of hydrogen wedge the rust away from the iron surface.
However, we should look at the electrochemistry to help understand what's happening. We will see that real-life results do not always depend on calculated reduction potentials.

The half-reaction

Fe³⁺ + 3e^- <----> Fe (-0.037 volts)

has a reduction potential that's more negative than

2H⁺ + 2e^- <----> H₂ (0.000 volts).

The half-reaction

Fe²⁺ + 2e^- <-----> Fe (-0.447 volts)

is even less favorable than Fe3+ into Fe.

On the other hand, the half-reaction

Fe³⁺ + e^- <----> Fe²⁺ (0.771 volts)

has a positive reduction potential, meaning it should be favored over all three of the above reactions.

Theoretically, the entire situation rearranges at high pH.
Fe²⁺ becomes Fe³⁺ (if O₂ is present), not the other way around. The reduction of Fe³⁺ to metallic iron becomes theoretically easier at higher pH, while it becomes more difficult for H⁺ to become hydrogen gas. The standard reduction potential for H+ into hydrogen gas, normally 0.000, becomes highly negative with increasing pH (perhaps because there simply aren't that many H+ around when the pH is high). By all accounts we should expect Fe³⁺ to reduce into metallic iron under such conditions. However, in practice it seems to happen to only a small extent, if at all. H₂ gas is still formed anyway, in preference to the other reactions.

It appears that the sparsely-available H⁺ (remember, this is alkaline pH we're talking about) turns into H₂ and vacates the system before ferric ion is reduced to iron.

There seems to be a widespread lack of understanding of basic electrochemical principles going around on the Internet. Some websites associated with electroplating or metal conservation have a good handle on the practical end of things, but they seem to have their principles backwards.

For example, at least a few web sites make the claim that reduction of Fe³⁺ to Fe is impossible, on the grounds that its standard electrode potential is -0.20 to -0.30 at pH 12-14. Actually, this potential is still not as negative as that of H⁺ ----> H₂ at the same pH (E = -0.60 to -0.80). One of the most basic principles of electrochemistry is that a reaction with a more negative reduction potential is less favorable.

Another important principle is that, if a species has a more negative reduction potential than Eq. 1 or 2 (below), it cannot be plated out on the cathode by applied EMF.
Eq. 1: 2H₂O + 2e^- <-----> H₂ + 2OH^- (-0.8277 V)
Eq 2: 2H⁺ + 2e^- <-----> H₂ (0.0000 V; decreases with increasing pH, down to about -0.80 V at pH 14)

For example, aluminum metal cannot be plated out on the cathode from aqueous solution, because the half-reaction

Al³⁺ + 3e^- <-----> 3Al

has a reduction potential of -1.662 volts. This is lower than that required to form H₂ gas; thus, water will decompose to form hydrogen before aluminum ions can plate out of solution as metallic Al.

The plating of certain metals, such as zinc, is often accomplished in alkaline baths, because the reduction potential of Eq. 2 (above) plummets in alkaline solution. This allows the plating-out of zinc metal to take precedence over the formation of hydrogen gas. In other words, at that pH the reduction potential of Zn²⁺ to Zn is less negative and therefore more favorable than that of H⁺ to H₂.

Now, let's go back to the possible reduction of ferric oxides to iron metal.

The reduction potential of

Fe³⁺ + 3e^- <----> Fe

is somewhere around -0.30 at pH 14, as mentioned above; this is less negative than the reduction potential of either Eq. 1 or 2 (above). However, a given Fe³⁺ ion has access to a great number of OH^- ions in alkaline solution; it may simply form a precipitate of insoluble ferric hydroxide before it has the chance to be reduced to iron metal. Alternatively, there may be other factors we haven't even considered. Nevertheless, reduction potential alone isn't sufficient reason to say that Fe³⁺ cannot turn into Fe in alkaline solution.

Supplies: Full safety goggles and / or face shield; Sodium carbonate (Washing soda); Calcium hydroxide solution ("Limewater")¹; Glass tubing; Distilled water; Brownlee electrolysis apparatus; Current-regulated power supply²; pH paper; Balance or scale with readability to 0.1 g or better
WARNING: Electrolysis mixes electricity with a liquid, something that's not normally advised unless you absolutely know what you're doing when it comes to voltage and current. NEVER use wall current for electrolysis. Also, if you do not know with 100% certainty what gases are created upon electrolysis of a given solution, do the procedure under a fume hood or outside.

General Procedure: Dissolve enough sodium carbonate in the distilled water to make a 1% solution (w/v). Also prepare a 2.5%, a 5%, and a 10% solution in separate containers. Don't forget the correction introduced by any water of hydration that's present; "washing soda" is Na₂CO₃·10H₂O (though it can lose some water of hydration at room temp). Calculate the pH that should result from making each of these solutions; or, get at least an estimate of the result by using pH paper.

Heat-bend a piece of glass tubing so that it can be run from the anode of the apparatus (the one at the "+" electrode) out of the jar and into a flask of limewater. In other words, instead of the gas at the anode bubbling up into a test tube, it will bubble up into the glass tubing where it can be delivered into a flask of limewater at the other end.

This setup will detect carbon dioxide if it is formed in any significant amounts at the anode. The test is positive if a white precipitate forms when the gas starts bubbling into the limewater solution; this white precipitate is calcium carbonate (CaCO₃). For discussion: is there anything that could come through that gas delivery tube which could give a false-positive "CO₂" result?

Run each solution in the electrolysis apparatus for the same amount of time. That is, try the 1%, the 2.5%, the 5%, and finally the 10%, cleaning the apparatus thoroughly each time. What happens to the pH after a fixed time of electrolysis in each case? What, if anything, happens to the rate of formation of carbon dioxide at the anode, as evidenced by the degree of cloudiness of the limewater? Where would you guess the carbon dioxide is coming from (propose a reaction)? Assuming there's an excess of limewater, the precipitated calcium carbonate could be collected, dried, and weighed. Working back with a little stoichiometry, this enables a rough quantitation of CO₂ whose imprecision would make an analytical chemist cringe³.

II. Becquerels to Sieverts... or, You Can't Get There From Here

Collectors of radioactive minerals have access to a variety of second-hand radiation detectors. Some of these date to the 1950's or 60's and were designed for nuclear fallout monitoring, while others are more recent surplus from labs where radioactive tracers such as ³²P were used. The scales on these meters can have many different unit markings: cps, cpm, mR/hr, mSv/hr, and more. The units can be confusing, since there's no easy way to convert from something like becquerels (Bq; a measure of nuclear transformations per second) to sieverts (Sv; a measure of biologically-absorbed dose equivalent). Do you have units in nanocuries but want to arrive at millirems? As the saying goes, you can't get there from here... at least, not exactly. Are these units of any use to the mineral collector, then?

Though there's no universal, direct conversion between the two types of units, there is help on this subject. The WISE Uranium Project has some useful Java-based calculators for determining uranium decay and doses. Let's try putting some values in that site's External Radiation Dose calculator. Suppose we have a uranium ore specimen which is 20% uranium by weight and approximately 3 cm deep and has a radius of 5 cm (this would be a very "hot" rock that would easily "peg" the needle on our counter); the Java-based calculator returns about 240 microsieverts per hour as the dose 1.0 centimeter from the sample. Directly against the skin, however, the dose increases to about 900 microsieverts (0.9 millisievert) per hour. A slightly thicker ore specimen (more uranium) would push the dose to around 1.2 mSv/hr, assuming it were held right against the skin for that whole duration (not a good practice!).

What about radiation meters, then, whose scales read "directly" in dose equivalent units, such as micro- or millisieverts per hour? Such a scale cannot be accurate for every kind of radioactive source. The units which characterize the sievert (Sv) are joules per kilogram (J/kg). Since joules describe energy, and since energy varies between radioisotopes, this must be true regardless of how reassuring those markings look on your meter's scale. A meter has to be calibrated based on the decay energies of whatever source it was designed for; a meter calibrated for ¹³⁷Cs or ³²P is not going to give correct dose-equivalent readings for ²³²Th or ²³⁸U. Some readers may notice problem with this example itself... ²³⁸U is in a different emitter class from ³²P, which in turn is different from ¹³⁷Cs. Alpha, beta, and gamma are their primary decay types, respectively; for more information, have a look in the Handbook of Chemistry and Physics by CRC Press. You'll see that ²³⁸U and ²³²Th decay by alpha emission.⁴

Despite these complications, a meter marked with units such as uSv / hr or mrem / hr is still useful to the mineral collector. While the meter's absolute numeric readings aren't necessarily correct, the relative readings are certainly useful. After a collector becomes familiar with what kinds of U and Th minerals make his or her meter give just a few audible clicks -- and which ones "peg the needle"-- it's easy to know which are the mild specimens... as opposed to which should be in that special display cabinet with the outdoor-exhausting fan and the 3/8" Plexiglas front....

III. An Interesting Trick with Equilibrium

In chemical equilibrium, we can think of a reaction as proceeding in both directions at the same time (or we can think of two simultaneous reactions which are opposites of one another). To use an example that seems to have been in every textbook since the dawn of modern chemistry: NO₂ (brown, gaseous) and N₂O₄ (colorless, gaseous) are in equilibrium at room temperature; some of the NO₂ is always forming N₂O₄, while some N₂O₄ is always converting back into NO₂. For the beginner, a question arises right here: does "equilibrium" mean "equal concentrations"? No, but it does mean equal rates of reaction. There could be 100 grams of NO₂ and a mere fraction of a gram of N₂O₄ at a given time⁵, but some NO₂ is forming N₂O₄ at a certain rate, and some N₂O₄ is changing back into NO₂ at this same rate in order to maintain the ratio under given conditions. Conversely, if the forward and reverse rates are not equal, a reaction is not in equilibrium.

A reaction in equilibrium has an equilibrium constant which essentially tells us the ratio of "reactants" to "products" at a certain temperature and pressure. While either side of the equation could be thought of as reactants or products (depending on which way you're looking), for sake of this discussion we'll use the "left" as "reactants" and the "right" as "products". No mnemonics here- just think of starting on the left and finishing on the right, even though there's really no starting and finishing in equilibrium. Assuming we could measure the exact concentrations at a particular temperature, we'd determine the equilibrium constant of the NO₂ <---> N₂O₄ system by finding equilibrium concentration of N₂O₄ and dividing this by equilibrium concentration of NO₂. This would be expressed as [N₂O₄]_gas / [NO₂]_gas .

Solubility is a type of equilibrium in which the aqueous ions can be thought of as "products". The undissolved substance could be thought of as a "reactant", but it does not figure into the equation for the special equilibrium constant called solubility product, or K_sp. Because it is not dissolved, a solid cannot have a "concentration". Only the dissolved components will show up. Thus, K_sp for barium sulfate is equal to [Ba⁺⁺]_aq times [SO₄^--]_aq; the brackets denote concentrations in moles per liter.

Even the most "insoluble" ionic solids will give a nonzero concentration of aqueous ions; it's just that they may be difficult to measure. The K_sp may be extremely small, but it is not zero. Consider barite (BaSO₄), commonly thought of as being "completely insoluble" in water; however, there is enough Ba⁺⁺ and SO₄^-- in solution at room temperature that we can make use of a simple trick to solubilize the barite without resorting to concentrated acids. While the amateur mineralogist will find fusion with powdered potassium carbonate or sodium carbonate to be quicker, this interesting technique is worth knowing.

Introduction to Semimicro Qualitative Analysis (Sorum, 1953) outlines the method. The underlying idea is to sequester or "grab" the barium ions that go into solution, taking them out of the equilibrium between solid BaSO₄ and [Ba⁺⁺][SO₄^--]_aqueous. What then happens is that more barium ions will go into solution to replace the lost ones.

In this example we use a saturated solution of potassium carbonate or sodium carbonate to supply CO₃^-- ions. Carbonate ions will combine with barium ions to give a precipitate of barium carbonate; the reason this is so useful is that barium sulfate is stubbornly insoluble in most acids (except concentrated sulfuric, and then only slowly), while barium carbonate will dissolve even in dilute HCl.

One may have a question at this point: if barium sulfate is so much less soluble than barium carbonate, why doesn't the sulfate "grab" the barium ions back from the carbonate and re-precipitate them as barium sulfate? This is normally what would happen; however, because we are using saturated potassium carbonate to supply our CO₃^--, there will be a huge excess of carbonate ions in the solution relative to the number of sulfate ions. Thus, it's much more likely that a Ba⁺⁺ will combine with a CO₃^-- than an SO₄^-- ion.⁶

To see if this is really the case, let's try an experiment. First, put on a good set of safety goggles. Next, carefully crush a small piece of barite into powder and put equal parts into two different test tubes. Fill tube #1 about one-fourth full with HCl; fill tube #2 about one-fourth full with potassium carbonate solution. Cover both of these and let stand for a week or two (better yet: a month or two), agitating periodically to re-suspend the barite powder. At the end of the time, make sure all the particles have settled and then decant the potassium carbonate solution from tube #2, replacing it with a small amount of dilute HCl. Any barium carbonate that has formed will dissolve in the HCl; the barium sulfate (barite) will be left behind. The longer the barite has been left in the potassium carbonate solution, the more BaSO₄ will [hopefully] have been replaced with BaCO₃. Do you find this to be the case?

Try a flame test with the liquid in each tube by dipping a clean platinum wire into the acid. Solution #1, our "control", should not give a green flame test as long as there are no solid particles of barite sticking to the wire. The solution in tube #2 should give a green flame test (we hope) and should also yield a white precipitate upon addition of excess sulfate ions. Update: the author's test solutions became contaminated with sodium ions from dust in the air over the course of a month or two, so the experiment has to be re-done to get definitive results.

Chemical equilibrium is actually a fascinating topic when one realizes the usefulness of it-- as well as how commonly equilibrium situations occur in all of chemistry and related sciences. The amateur mineralogist can certainly benefit from a basic understanding of equilibrium and the chemical "tricks" that can be done with it.

IV. Site News

We're pleased to announce the arrival of the new "Tachometer" and "Select-speed" versions of the Ultra 8 centrifuge. The Ultra 8 "Select" model has four settings corresponding to CLIA-recommended speeds for standard clinical centrifugation of body fluids. The Ultra 8 "Tachometer" model has variable speed and a built-in tachometer with digital readout-- useful for protein purifications and any other experiment where it's important to know and control the exact relative centrifugal force (RCF). If you have an original Ultra 8V or other centrifuge without built-in speed indicator, we now also sell a hand-held tachometer.

More glassware is now available in our on-line catalog, including 500 mL distilling apparatus and the elusive glass retort in 250 and 500 ML sizes. There are also more types and sizes of clamps, stands, and support rings now available.

That concludes this issue of the CR-Scientific newsletter.
Until next time, stay safe and have fun.

Email: sales_AT_crscientific_DOT_com

(replace the "_AT_" with an @ symbol and the "_DOT_" with a period. This is to thwart spam-bots.)

Notes:

¹ Limewater can be prepared by dissolving slaked lime (Ca(OH)₂) in water. Quicklime (CaO) will also work, but it will generate considerable heat at first. Either way, treat the solution with care- it is caustic and can cause skin burns and permanent eye damage.

²if a current-regulated power supply is not available, a 12 volt battery can be used ONLY if a suitable resistor is placed in series with the circuit. For example, a 100 ohm resistor will allow at most 0.12 ampere (120 milliamperes) of current to flow at 12 volts; the resistor should be rated at 2 watts or better in order to dissipate the heat that will result. NEVER use a car battery or other large battery UNLESS also using a suitable resistor to limit the current. It is imperative that you be proficient with V=IR and P=I²R before working with electricity.

³ There were more reliable ways, even 75-100 years ago, to quantitate CO₂. While it's very easy, the crude method outlined here does not have any means of removing unwanted gases from the CO₂ that's generated. How do we know that CO₂ is the only gas produced that will make a precipitate with Ca(OH)₂ solution? Traditional analytical methods use anhydrous H₂SO₄, anhydrous CuSO₄, or other sorbents to remove unwanted gases such as H₂S from the carbon dioxide. There's also the matter of how pure the Ca(OH)₂ is or isn't.

⁴ Many detectors cannot even sense alpha, since a special thin-window tube is necessary to admit these easily-stopped particles. A simple way to find out if your detector is picking up alpha is to interpose a sheet of paper between the ore specimen and the G-M tube. Does the needle drop at all? This will work only if the distances are short enough that the air isn't absorbing the alpha before it can get to the G-M tube (!); the specimen should be about a centimeter from the window in the end of the tube. Even if your detector does not register alpha at all, you've probably noticed it has no problem detecting radioactivity from uranium and thorium ores. This is because some of ²³⁸U's and ²³²Th's decay progeny are beta and gamma emitters.

⁵ this arbitrarily-selected number doesn't at all take into account what the real equilibrium concentrations of these two gases might be under normal conditions. The dark reddish-brown NO₂ is greatly favored over N₂O₄ at higher temperatures.

⁶ The example in Sorum's book uses lead sulfate, which is converted to the even less-soluble lead carbonate. Barium carbonate is more soluble than barium sulfate. However, the excess of carbonate ions can drive the precipitation toward one less favored (i.e., barium carbonate which has a higher K_sp). If the sulfate ion concentration started to become large relative to the carbonate concentration, the "trick" would stop working in the case of barite.

Works Cited / Suggested Reading:

Hillebrand, W., and Lundell, G. Applied Inorganic Analysis. New York: John Wiley and Sons, 1929.

Sorum, C.H. Introduction to Semimicro Qualitative Analysis. New York: Prentice Hall, 1953.

Weast, Robert, ed. CRC Handbook of Chemistry and Physics. Boca Raton, Florida: CRC Press Inc., 1988.

While the information in this newsletter is thought to be accurate to the best of the authors' current knowledge, it is not guaranteed to be free of errors or to be suitable for any particular use. The procedures and experiments outlined within can be dangerous or even fatal if carried out improperly. If you choose to attempt any of them, you proceed entirely at your own risk.

This newsletter is copyright of CR-Scientific, 2003. You may distribute it freely provided the contents of the file are not truncated or altered in any way. Please email the address given above if you find any errors or omissions or would simply like to make a suggestion.

General Terms of Use for this website

CR-Scientific Catalog
Newsletter Index Page
Lab Glassware
Main Page