Cleaning and Polishing Brass (III)

Oxidation | Cleaning | Polishing | Ammonia | Cyanide
Effects of Quality | Stress Corrosion Cracking
Commercial Products | Electrolysis

Effects of Quality:
When impurities in metals are found in higher concentrations in some areas and in lower concentrations in other areas, the difference in electrical potential between the metal atoms and the atoms of the impurities would promote oxidation whenever an electrolyte (salt bridge) is present on the surface of the metal, forming tiny voltaic cells on the surface. These cells can work in series and accelerate the oxidation process even more. If the impurities could be distributed perfectly evenly in the metals, the uniform distribution would minimize the collective effect (voltaic cells working in series) and thereby minimize the rate of oxidation. Because of considerable improvements in the production methods of metals, modern metals (particularly since WWII) are much more homogenous (the impurities are more evenly distributed in the metals) and have fewer impurities. Modern metals will oxidize less quickly than metals made by older production methods. The differences can be seen in antique and modern clocks.
Production methods in antique clocks frequently including pounding, beating, and hammering of the brass to the desired shape and thickness, resulting in hardening of the brass and in stresses. The stresses tend to cause stress cracking in the metal on a microscopic level at the points of highest stress and of greatest weakness in the metal structure. Very serious stress cracking in poorly fabricated metal can sometimes be visible to the eye. The microscopic stress cracking is revealed when the oxide layer is removed. Note that the use of ammonia or even of acid for a few minutes to remove the oxide layer does not cause stress cracking. The use of these chemicals merely reveals pre-existing defects in the metals. Since the amount of metal removed is very small indeed (in brass, perhaps one tenth of one percent), the harm done by the removal is insignificant. A purist, however, would not want to do this to a valuable antique timepiece!
A modern manufacturing technique used to evenly distribute the stresses in brass clock plates can be seen on many Hermle clocks, which have a pattern stamped onto the plates. Brass obtained from large rolls of sheet metal has a tendency to return to its curved shape, giving the plates a tendency to warp: the pattern that is stamped onto the plates stretches the brass and removes the tendency to warp so that the plates remain flat. Combine this technique with modern methods of manufacturing brass, resulting in much lower levels of impurities that are much more evenly distributed throughout the metal, and the end result is a product of extremely consistent quality that the manufacturer can control very precisely.
Tempered steel that has an unequal distribution of temper (in other words, that was not tempered evenly) and that has an uneven distribution of impurities is similarly vulnerable to stress cracking, which is usually what causes mainsprings to break.

Stress Corrosion Cracking:
A much more serious form of stress cracking could occur if the brass were exposed to an oxidizing agent or an acidic environment with the presence of ammonia or ammonium salts, because they form complex ions with zinc and copper, namely tetraaminezinc   [Zn(NH3)4]++  and tetraaminecopper   [Cu(NH3)4(H2O)2]++   ions (which gives the acqueous solution a blue colour).
The protective layers of zinc and copper oxides, which protect the metal underneath because they are insoluble in water, become soluble with the formation of the complex ions and are removed from the surface of the metal, exposing fresh metal to be oxidized. The oxidation continues unimpeded until the metal dissolves. If a stress is applied to the brass (such as hanging a weight from a brass hook), the brass will crack at the points of greatest stress when enough metal is oxidized to cause the lattice structure of the metal to fail. For example, if you hang a lamp from a brass hook in a barn (where there is plenty of moist air, carbon dioxide and ammonia produced by the animals), the moist air results in the formation of a layer of electrolyte (weak carbonic acid from carbon dioxide dissolved in water) and the right conditions are created for corrosion to take place, unimpeded by any protective layers of insoluble oxides or hydroxides as the ammonia makes the metal ions soluble. The warmth that rises from the lamp below accelerates the corrosion, the hook cracks and breaks, allowing the lamp to fall. The cracking that takes place here is called "Stress Corrosion Cracking" because this cracking is a direct result of corrosion. This cracking takes place when a stress is applied, but not if a stress is not applied: the brass hook will not crack if the lamp is not hanging from it! The brass hook will corrode, though.

Note, however, that the clock cleaning solution does not have any acidity (hydrogen ions in solution) nor any oxidizing agents because the solution is alkaline and ammonium hydroxide is a reducing agent. Since metals do not oxidize in alkaline solutions, the ammonia present does not dissolve the metal atoms underneath the layer of metal ions that form the metal oxides. The layer of oxide is removed and bare metal is exposed, but no corrosion takes place in the solution. No corrosion takes place until the metal is removed from the solution and exposed to air. There are no applied stresses either (unless you submerge the entire clock movement with the mainsprings wound up: anyone who does this should promptly depart from this profession anyway) since the clock should be disassembled prior to cleaning. The necessary conditions are not met when the brass is submerged in ammoniated clock cleaning solution for stress corrosion cracking to take place, so stress corrosion cracking does not happen here. In the equations above, the zinc and copper oxides become tetraaminezinc acetate and tetraaminecopper acetate.

Commercial Products:
Commercially-prepared ammoniated solutions for cleaning clocks are more sophisticated. The traditional formula includes ammonium hydroxide solution, oleic acid and acetone. The acetone is a powerful organic solvent that is also soluble in water because the oxygen atom is polarized and forms a weak bond with the hydrogen atoms of water molecules. As an organic solvent, acetone aids in the removal of dirt and old oils from the clock parts and in dissolving the oleic acid. The oleic acid is made from maize (corn) and is basically an alkene with a long carbon chain and a carboxylic acid group at one end.

which could be represented as: R-COOH

Oleic acid is a liquid at room temperature (at least in Texas)because it is an unsaturated fatty acid. (For comparison, stearicacid, which also has a carbon chain with eighteen carbon atoms, isa saturated fatty acid, so it has a higher melting point. Stearic acidis made from animal and vegetable fats.) Oleic acid is a weak acidlike acetic acid because it ionizes very little in aqueous solution,but it does not dissolve in water the way acetic acid does, forminginstead a suspension of groups of organic ions called Micelles,thereby behaving like a soap. Oleic acid reacts with ammoniumhydroxide to produce ammonium soap:

This reacts with metal oxides to produce metallic soaps:

The oleic acid therefore has two functions: to remove the oxide layer from the metals and to act as a soap to remove the dirt and oils. The cleaning solution has about 90% ammonia solution with about 5% oleic acid and 5% acetone added, thereby leaving plenty of excess ammonia molecules to keep the solution alkaline.

The latest ammoniated cleaning solutions take advantage of detergent technologies. The oleic acid is replaced with detergents because the detergents are not severely affected by hard waters with calcium and magnesium ions.

I include the following quote verbatim from information Matthew Headrick provided me because his choice of words is excellent: "Ammonia, like alcohol and acetone, is also valuable in cleaning solutions because it is highly volatile, so the residual cleaning solution evaporates much faster off the cleaned surface than water alone would, preventing streaking, further oxidation, and other bad things." Therefore, there should be no fear about any ammonia residue being left behind in cast brass, which is porous, ammonia which you might fear would react with the metals after cleaning.

The reason why household ammonia effectively replaced myammoniated clock cleaning solution that I recycled (see previousessay) is because the reactions are essentially the same. The oleicacid is replaced by several different kinds of detergents (for a widevariety of grease-cutting applications), referred to on the label asanionic and nonionic surfactants. The oxide layer is similarlyremoved.

If you wanted to polish the brass plates despite the removal ofmetal atoms you could buy household ammonia from yoursupermarket. It contains about 5% aqueous ammonia, a weakalkaline, and very effective surfactants. Do not buy any cleaners that do not list the ingredients on the container. You could add a smallamount of acetone (about 5% by volume) from the same businesswhere you buy paints for your house. But please: if you do notunderstand the chemistry outlined here, buy the cleaning solutionsprovided by your clock parts suppliers. If you do not know whatyou are doing, you could easily ruin an antique clock!

My recommendation is to clean the clock parts by hand with anorganic solvent, such as tetrachloroethylene, but be careful not toinhale the fumes from the solvent, which will make you ill. Do notremove the protective oxide layer from the brass, but do removethe oxide layer off steel parts. If you prefer not to use organicsolvents because of the fumes, use the latest cleaners anddegreasers with detergents without ammonia and without acids, butthese water-based cleaners require that the clock parts bethoroughly rinsed in water and dried immediately to prevent theformation of new oxides. I dry my clock parts with an aircompressor at about sixty pounds per square inch (in the tank) verythoroughly and then in a metal box with four light bulbs (60w.) thatwarm the clock parts for a couple of hours to evaporate any remaining traces of water.

From a theoretical point of view, it would be possible to clean the oxide layer without removing metals atoms from the surfaces if the clock parts had their oxides reduced by electrolysis. If the part were attached to the cathode and immersed into dilute sodium hydroxide solution and a potential difference of 4.5 volts (which should be enough) of direct current were applied for a couple of minutes, the metal oxides would be reduced to the metals and the oxygen ions would float away into the solution. Prolonged electrolysis would result in the gradual disappearance of the carbon anode. In my opinion, electrolysis is a lot of effort to go to only to save a few hundred metal atoms that protected the metal atoms underneath (which are then exposed to oxidation themselves). However, it is interesting to know that the removal of oxides from brass and steel parts can be achieved without the loss of metals: this should make you think of the Hall-Heroult Process, by which aluminium metal is extracted from its ore.


"Structural and Comparative Inorganic Chemistry" by P.R.S Murray and P.R.Dawson. 2nd reprint, 1980. This was one of my A level Chemistry textbooks from Sedbergh School.

"Chemistry" by Dr. Steven Zumdahl, University of Illinois, third edition, 1993.

"Chemistry made Simple" by Dr. Fred Hess, revised edition, 1984.

Allan Headrick, B.S. (Swarthmore, Physics). Teaches chemistry and physics and Austin High School (Austin, Texas). (Allan is my brother). Allan provided most of the information that formed the basis of this essay.

Dr. Steve Herron, B.S. (Heidelberg College, Ohio), Ph.D. (Iowa State University, Organic Chemistry), of Houston, Texas. Steve is the son of Sue Wysong, a local watchmaker. Steve answered my questions in a one-hour telephone conversation. Steve also provided much-needed help with information about complex ions of copper and zinc.

Matthew Headrick, B.S. (Princeton), graduate student at Harvard University, working on his Ph.D. in Physics. Mat gave me some important information about solvents and about the action of soaps on lipids. (Matthew is my cousin.)

Go to Russian Clock

Clock Repair Main Page
Escapements in Motion
Links Page