Electrolytic corrosion

Author: 
Mike Harris

Fitting new seacocks, skin fittings, rudder pintles, stern gear or any underwater metalwork can be a trying task, what with all the trouble of hauling out and then having to jam yourself into those parts of the bilge that normally shaped people just can't reach. The work itself isn't much fun either. With the all pervading dust and filth its never a pleasant job but none the less has to be conducted with great care as the watertight integrity of the hull depends upon it.

However, even when you have taken great pains over the work and used the best fittings your hard earned money can buy, it can be sickening to find at the next haul-out that the metal has become deeply pitted or looks spongy. Electrically assisted (ie. electrolytic or 'galvanic') corrosion is the likely cause and it seems rather unfair that good quality marine grade materials and honest workmanship are no guarantee that it will not occur.

A better defence lies in an understanding of the circumstances that promote electrolytic corrosion and then, armed with this enlightenment, you should be in a position to control its effects on your own boat. In this article we will take a brief look at the theoretical causes of electrolytic corrosion. In a laboratory, such ideas are quite easy to demonstrate but in practice on a boat, things are more complex and any tidy theory is often upset by many factors that can be difficult to predict or are beyond our control. Because of this it is often revealing to see how the theory works out in practice so we will also take a look at a few practical examples of corrosion difficulties.

Basic ideas
Before electrolytic corrosion can take place the first thing that is needed is an electrolyte. Any acid, alkali or salt solution will do but sea water is of course the most common electrolyte and the one we are mainly concerned with here. When a metal is immersed in such a solution there will be a tendency for some to dissolve away. With metals such as silver or gold, even after many years of immersion the loss is negligible, but most other metals are more reactive and corrosion is more likely. In practice, the chemical nature of the metal and electrolyte are not the only factors that determine the actual amount of material that a metal structure will lose due to corrosion.

The presence of local electrical fields can also have a profound effect and these are established whenever electrons flow from one part of the metal to another. This they can do most easily and quickly, and such flows are more familiarly known as electric currents. Even without connection to a battery or external electrical supply, the corrosion process will generate its own source of electricity with the circuit flowing partly through the metal and partly through the electrolyte. Current in the circuit may be tiny (a few tens of milliamperes) but without this no corrosion can take place, so how is it produced and what steps can we take to control it?

By their nature, metals towards the bottom of table 1 have a tendency to withdraw electrons from others higher in the table. For this to occur both must be in electrical contact and immersed in an electrolyte solution, but electron movements through the metals is only one part of the story. To complete the circuit there is a different kind of conduction carried out through the electrolyte.

Here, electric currents are carried by ions rather than electrons, and the type that are of most interest to us are formed from anode metal atoms. Normally these have no electrical charge but if any of their normal compliment of electrons are lost, they turn into positive ions. In comparison with electrons, ions are large slow moving cumbersome objects, but because they would really prefer to be neutral once again, they become active and leave the anode surface to hunt around for some negatively charged particle to react with. This they find by drifting off into the electrolyte where they may form intermediate compounds but their exact fate will depend upon which metals are involved. In any event the circuit is completed by the recovery of electrons at the surface of the cathode. See Fig 1 below.

Corrosion in practice
So much for the electrical circuit associated with electrolytic corrosion but, as with any other electrical circuit, break it anywhere and you will stop the action. One method is to insulate the metal parts either from each other or from the electrolyte. Plastic insulating washers placed between aluminium window frames and a steel hull are one example and a good paint system can also provide a successful barrier particularly on parts that are not permanently immersed.

Below the waterline it is a good idea to try to avoid combinations of dissimilar metals wherever possible. On my own boat which is made of steel I decided not to have bronze skin fittings for toilet, sink or bilge outlets. Instead I welded short lengths of threaded steel pipe directly to the hull and used polypropylene seacocks. (see (a) below). This eliminates the need for bronze or gun metal seacocks and skin fittings but there are other conventional uses of these metals that are also possible to avoid on steel boats.

Propeller shaft (cutless - or should that be cutlas) bearings are normally contained within a bronze housing. Some manufacturers supply these in phenolic plastic housings which of course causes no electrolytic problems. Holders for these bearings and the 'P' or 'A' brackets sometimes used to support drive shafts are also frequently made from bronze but with metal hulls there is no reason why you should not fabricate them from the same material as the hull and so avoid trouble from this source.

Of course when it comes to materials for the shaft itself and the propeller, there are few alternatives to the use of bronze and stainless steel, but if you were able to use the same metal for every metal part below the waterline would the problems of electrolytic corrosion then go away? Sadly they would not and there are several reasons for this.

Firstly any electric current flowing through the hull material can have a profound effect upon the rate of corrosion. These need not necessarily originate from any corrosive reaction but perhaps come from some equipment that may have been connected to the hull, and their effect will be to cause one region of metal to become more anodic and so corrode more quickly than another. Also, where corrosion is already taking place, they may combine with the current that it produces and so increase (or possibly decrease) the rate at which metal is lost. For this reason it is always a bad idea to use the boat's metalwork as the return connection for any electrical equipment.

Another reason is that metalwork in contact with sea water containing high levels of dissolved oxygen is likely to be more anodic than contact areas where oxygen levels are lower. Water, disturbed by winds and waves near the surface will contain large amounts oxygen and this effect helps explain why waterlines often suffer more corrosion than other parts of the hull.

A further point is that it is more than likely that the sheet material used for hull plating will be of a slightly different composition to the weld metal used to join it them together. Also, though one piece of steel, aluminium or bronze may look much the same as any other piece of the same metal, it is quite likely that invisible differences in composition or physical structure may exist from one region to another. The differences may be microscopic but in the presence of sea water the dissimilarity could be sufficient to form the basis of an electrolytic cell and the start of a corrosive reaction. This helps explain why rust corrosion on steel plates tends to proceed more rapidly in some areas, forming deep pits, rather than at an even rate over the entire surface of the metal. Common brass; a mixture of copper and zinc, is another example. When in contact with sea water the copper tends to promote the release of zinc ions from the remainder of the metal. This causes the part to slowly disintegrate and is the reason why brass screws, pipe fittings, gate valves etc are not a good choice for underwater fittings.

Protective anodes
Perhaps the best known way of preventing underwater electrolytic corrosion is through the use of protective anodes. They work on the idea that if you have a piece of metal attached to the hull that is higher in table 1 than any other metalwork, (ie more anodic) it will corrode away in preference to other hull parts.

Zinc is the material most commonly used but the composition of zinc anodes is carefully controlled since the presence of small quantities of other metals can have a large influence on their performance.

To gain maximum effect, anodes need to be mounted in close proximity and soundly connected to the fittings they are intended to protect. In the case of wooden or glassfibre hulls this generally involves joining them on the inside of the hull with a heavy copper cable (6 to 10sqmm). With steel boats, as the hull material is itself conductive, it is usually sufficient to weld or bolt the anodes to studs that are themselves welded to the hull. (see (b) below)

Can you fit too many anodes?
Whilst zinc anodes are corroding away they are providing a current which opposes any corrosion current that might be produced by the metal parts that we are trying to protect. Obviously, once the anode has wasted away, it can no longer do its job but a greater working life will be obtained by using larger or more numerous anodes. However, in some circumstances this may not be quite such a good idea.

As zinc anodes decompose, one of the products formed is an alkaline substance that is usually washed away in the surrounding water and has no effect on the hull. However, on wooden hulls, localised damage can occur where the concentration of this substance is high. This often shows up around the bolts used to secure the anodes which may appear to loosen as the wood around them softens.

It could also be that, in some cases, overdoing the number of anodes may encourage bronze parts to foul up with marine growths:

George Taylor (retired editor of the magazine 'Practical Boat Owner' observes that bronze cannons from ancient warships are often recovered after centuries of submersion in excellent condition and free from marine growth. However when bronze propellers are protected by large numbers of zincs, they frequently grow large amounts of weed and barnacles. If anodes are not used propellers can rot away but remain clear of marine growths.

The most likely explanation for this is that when left to its own devices a small amount of copper is leached from the bronze which discourages marine growths. By using large amounts of zinc, this leaching is reduced and so barnacles, weeds etc. can take hold, but there is a delicate balance to be obtained between using just enough anodes to balance out any corrosion currents without blanking off just sufficient copper release to deter fouling.

With new boats, electrical fields in and around the hull are never totally predictable. Also, if you have recently made changes to the underwater fittings or fitted some piece of equipment (a new engine for example), small changes in the underwater electrical environment may have taken place and electrolytic corrosion could suddenly become a problem. In all of these cases it is a good idea to keep a close watch on the situation, especially throughout the first year, so that any possible corrosion sites can be detected before they become expensive to correct or the safety of the vessel is threatened.

Electrolytic corrosion on parts above the waterline
Electrolytic corrosion does not only occur on underwater parts of boat hulls but can affect any dissimilar metal parts where only a tiny amount of moisture is needed to provide the electrolyte needed to set up a small corrosion cell.

One of the most common examples of this effect occurs when stainless steel bolts are screwed into aluminium threads, or are used to secure aluminium fittings. Months or even years later there may be no external signs of trouble but if any water has got into the threads, when you come to remove the bolt you may find it seized. More persistent attempts at removal are likely to strip the threads and reveal that much of the metal surrounding them has become changed into a crumbly mass of white powder.

In these applications, monel (an alloy of 70% nickel and 30% copper) is a better choice than stainless steel but unfortunately it is not easily available other than in pop rivets. Another approach would be to try to exclude moisture from the joint with sealing compound or to insulate the bolt from the fitting it is securing.

Oxide protection
With some metals, (aluminium, stainless steel and lead are examples) their first line of defence against corrosion lies in their natural ability to form a protective oxide layer covering their entire surface. This is formed by the reaction of the metal with oxygen from the air and forms spontaneously whenever they are cut or shaped. This formation explains why Aluminium, which from table 1 forms the basis for corrosion resistant alloys that perform much better than metals such as mild steel or cast iron that are much lower in the table but which do not readily form such inert oxides.

The finish given to 'anodised' aluminium parts such as masts, spars, cleats, window frames etc, is simply an electrical process that increases the thickness of this natural oxide layer. Not only does this improve the part's resistance to corrosion but is also harder than the underlying metal and helps improve the overall appearance.

Whilst I was still at school, I was frequently inspired with ideas from the magazine "Practical Mechanics". The magazine is now long gone but I well remember one article where the author was describing the fabrication of a paint container. Aluminium was the chosen material and he recommended joining it with ordinary soft solder and flux. I thought this was impossible but read that the secret was to create a pool of flux and molten solder on the metal surface then to expose the metal and allow the solder to wet the surface by scratching through the pool with the edge of an old hacksaw blade. In disbelief I tried it and it worked. I was amazed but it was not an easy process and I can see why the idea didn't catch on. However, this example does show how different a material aluminium is just beneath the surface.

Crevice corrosion on stainless steels
Stainless steels are another group of metals with an oxide coat that helps them resist both electrolytic and chemical corrosion.

Many grades are available but the two most commonly encountered are 18/8 (containing 18% chrome and 8% Nickel) AISI304 and the marine grades eg. 18/10/3 (containing 18% chrome 10% Nickel and 3% molybdenum), AISI316 or EN38J. However there is a particular type of corrosion - crevice corrosion, from which all may suffer, though the marine grades are less susceptible.

This takes place on surfaces that are not exposed to oxygen and typical regions where this could occur are behind washers or where the threaded section of a bolt passes through a nut. In these cases the extent of the damage can only be revealed by dismantling the fitting. This is because all normally visible parts will probably be exposed to the air or to oxygenated sea water if below the water line and so may appear in perfect condition. A sudden failure could be the only warning that something is wrong and for this reason stainless steel is always a dubious material to use for underwater fittings.

In spite of this there are one or two regions where stainless steel is often used underwater - propeller shafts and rudder bearings for example, and I must admit that my boat is one of these. Unfortunately, there are no smart answers to all the questions on electrolytic corrosion, but being aware of how the rot might strike is of help if only to let you know what to look for on the next haul out.

 

Corrosion samples scrap book

So much for the theory. In the remainder of this article we'll look at some practical examples

 

This zinc anode has been reduced to a small fraction of its former size though it had only been in use for a short period. It has obviously been doing a good job but as it gets smaller its effect will be reduced. In this particular case the reason for this high rate of decay is probably due to stray currents from the boat's electrical system.

Weed and barnacles will certainly affect the performance of this boat but could the zinc shaft anode be the reason for their presence? The anode is there to protect the propeller but under other circumstances bronze parts leach just enough copper ions to inhibit such growths and so stay clean even after many years of submersion.

The use of bronze skin fittings on steel boats may set up a corrosive reaction in the immediate vicinity. In this case a length of steel pipe was welded to the hull in lieu of a skin fitting but still the problem persists. The problem could be caused by the nature of the waste material flowing out of the fitting, or slight differences in the materials of the hull, tube or weld metal. In any case, a solution could be to use a good epoxy paint system to insulate the area from surrounding sea water, but to ensure good adhesion this would require local grit blasting

Damage to paint work around the waterline is often caused by collision with floating debris. This can form a starting point for electrolytic corrosion which is encouraged by high levels of dissolved oxygen in surface water. On this boat the waterline steel work is deeply pitted but in this case the owner feels it is not enough to cause concern as the hull material is 6mm thick.

Last year I replaced the diaphragm and seals on this pump, gave it a quick test to make sure all worked OK, then put it away for use in an emergency. My mistake was not to wash it through with fresh water and dry it out before it was stored. Because some sea water remained inside, an electrolytic reaction was able to take place between the aluminium body and stainless steel screws that hold the diaphragm and valves in place. The result was several stripped threads in the aluminium and the white deposit that can be seen covering these internal parts. Pumps with plastic bodies avoid these difficulties.

Here is an example of a situation where electrolytic corrosion looks certain to take over but it has not happened. The window frame is aluminium, the screws are stainless steel, the nuts are nickel plated brass, the hull material is zinc sprayed steel and the whole lot has been in place for 15 years with no problems. I am sure that if sea water were to leak between the hull and frame, all kinds of reaction would take place. Whilst I should not care to recommend the combination its success depends upon the insulation provided by the aluminium anodising on the frame, the epoxy paint applied to the steel work and the integrity of the polyurethane sealant used between the hull and frame.

Crevice corrosion is an insidious kind of rot that affects stainless steel and sometimes takes place where you least expect it. The head of this bolt is in excellent condition and its markings indicate that its made of a marine grade of stainless (A4). For two years it was in service aboard a Danish schooner where it was well above the water line and helped secure the chain plates to the Larch hull. Quite clearly it is extensively corroded but this would have gone unnoticed until a failure had occurred or, as in this case, the fastidious owner had been checking it out during a routine inspection.

Corrosion cells can be set up between the wedge interface that forms between say a bronze shackle and stainless thimble. Electrical contact through the bearing surfaces is assured but the wedge also traps drops of water that lie there for long periods. This does not need to be sea water as once salts are present, they just re-dissolve in whatever other water comes along.
Using longer lengths of steel pipe in lieu of skin fittings so the seacocks lie above the waterline. this eliminates the criticism that in a fire, plastic seacocks can melt, thus letting in water that could sink the boat.
Attaching anodes to the hull. Welding them on directly would ensure the best electrical connection but would also destroy any surrounding paintwork on the inside of the hull. To avoid this some people weld short metal tags to the hull (say 25mm X 75mm) to which they then bolt or weld the anodes. In this way, the anodes are spaced away from the hull thus increasing their effective surface area.
Connecting metalwork together brings it to the same electrical potential. In the same way brushes can used make an electrical connection to the propeller shaft.
Marinised car engines are a popular alternative to the usually more expensive purpose built marine unit. Non-marine accessories such as alternators and starter motors are cheap but use the engine block as the earth connection. The circulating currents that this may introduce carry some risk of corrosion that can be eliminated by insulating the engine from the hull. Flexible engine mounts may alone be sufficient but don't forget the throttle, gear and fuel line connections.

� Mike Harris March 2005