🚀 go-pugleaf

In article <3167@sunny2.che.clarkson.edu>, kweeder@sunny2.che.clarkson.edu (Jim Kweeder) writes:
> Ok, Larry.  You want flames, you got them.

	I must be blessed.  Unfortunately, I am the type of person who does
not easily say "the hell with it" and give up - so you've still got my
attention and my time.

> >> >	The important point to bear in mind is that there is no significant
> >> >presence of dissimilar metallic junctions beneath an automobile which lend
> >> >themselves to the formation of galvanic corrosion cells which affect any
> >> >significant area of the automobile body
>  [Material Deleted]
> >	I already mentioned this in my previous two articles, in addition
> >to the above.  The point to bear in mind is that these other corrosion
> >mechanisms are NOT SIGNIFICANT in the automobile environment!
> 
> If we read these statements closely enough, we will find that automobiles
> never corrode.  Isn't that nice.  However, Larry, there are a number of
> upstate NY rust buckets in the parking lot of this building.  How did
> they get that way?  Gemlins?  Divine intervention?  Sorry, Larry, but
> you can't have it *both* ways.  I'm afraid that you're going to find
> out if you bother to look at any texts or design literature that
> concentration cells, and other non-dissimilar metal corrosion mechanisms
> are recognized as a problem in mild steel.

	Have I denied or disputed that automobiles rust?

	I have been trying to tell you two things:

1.	The corrosion mechanisms which are NOT, in general, responsible
	for rust in automobiles.

2.	How cathodic and anodic protection schemes are, in general NOT
	effective as an "aftermarket" measure for autombiles.

	So, I'm going to try to get through to you using another approach.

	Most textbooks define ALL corrosion of metal as being electrochemical
in nature.  While strictly speaking this is true, it may mislead an otherwise
intelligent but inexperienced soul - in this case, YOU - into believing that
with a sacrificial anode here and some impressed potential there one can stop
all of the corrosion in the world in one fell swoop.

	Wrong.

	Lets say that I take a 4 by 8 foot sheet of unplated, uncoated steel
and just set it on a couple of tires which are laying flat upon the ground.
For all intents and purposes, there is no electrically conductive path between
this metal plate and the earth.

	The above analogy is not unlike that of an automobile.

	Now, lets throw some water containing sodium chloride on top of the
plate, and look at the chemistry (the sodium and chlorine ions do not
participate here):

	2Fe + 2H2O + O2 --> Fe+2 + 4OH- --> 2Fe(OH)2

	What we have here is ferrous hydroxide, which precipitates, but
which upon exposure to oxygen and water is unstable and undergoes the
following reaction to form ferric hydroxde:

	2Fe(OH)2 + H2O + 1/2O2 --> 2 Fe(OH)3

	Ferric hydroxide, when it dries, forms what is commonly known as
rust.

	Now, the $64 Question is: from a PRACTICABLE standpoint, how can we
stop this rust?

	Well, we can do a number of things, including but not limited to:

1.	Painting the steel, other otherwise coating it with a substance
	which is impervious to water.

2.	Electroplating the steel with chromium, nickel, zinc, etc.

3.	Hot dipping the steel with zinc (galvanizing), tin, lead, etc.

4.	Cladding the steel with thin metal which is pressed in place using
	rollers.

5.	Using more exotic deposition processus such as flame spraying,
	vacuum deposition, sputtering, etc.

6.	Using a chemical conversion process, such as treatment with
	phosphoric acid (called phospatizing), which is sometimes used as
	a precusor for automotive body metal finishing.

	From a PRACTICABLE standpoint, that's about it.

	Now, I just know what YOU are getting ready to shout: "Sacrificial
anode to the rescue!"  "Impressed potential to the rescue!"  Okay, well let's
consider these alternatives.

	In order to apply cathodic protection that provides a reasonable
degree of corrosion retardation, it will be necessary to have a current
density of at least 5 mA per square foot.  So, let's take a zinc bar that
is say, .1 in thick, 1 inch wide and 8 feet long and run it down the
middle of this steel plate, and fasten it with a low-impedance connection
to the plate.

	Now, we have just reduced the steel surface area of 4,608 in^2 by
96 in^2 occupied by the sacrificial anode.  Okay, so how far away from
this one inch strip do you think we will have a current density of at
least 5 mA/in^2?  2 feet away (to the edge of the plate)?  1 foot away?
6 inches away?

	How about less than *** 2 *** inches away?  Now, I admit that for
the benefit of this article (you're killing my spare time doing even this
much) I have not done the calculations that involve the half-cell potential,
cell plate surface area, electrolyte conductivity, steel sheet resistivity,
and current/potential gradients, but from past experience I can tell you
that the answer is well under 2 inches.

	So, in order to have any reasonable chance at achieving cathodic
protection using sacrificial anodes, you are going to have to install a
zinc strip no farther than every 2 inches down the length of this sheet.
That's going to require at least SIXTEEN strips of zinc for this sheet,
with the strips themselves occupying ONE-THIRD of the surface area of
this steel sheet.

	Okay, so the above scheme of cathodic protection may work, but
be honest now, do you think it is PRACTICABLE?  Hell, you just may as
well clad the entire surface with zinc and be done with it!

	But do you realize something?  If you clad the entire surface with
zinc, you no longer have an iron-zinc couple which is exposed to the
electrolyte.  So, what you really have is NO LONGER CATHODIC PROTECTION,
afterall.  You will not have cathodic protection in any area other than
where the zinc surface is broken.

	Do you understand, yet?

	Okay, now let's cover the issue of impressed current from a power
supply for cathodic protection.  Well, we have a similar problem with the
sacrificial anode example above: how to get enough current density.  In
the case of an impressed current cathodic protection system, the anode
will have to be _insulated_ from the cathode (i.e., the steel sheet), which
will significantly complicate the installation.  While we can now increase
the anode spacing and reduce the surface area, we are not under these
conditions of a plate with a THIN electrolyte film going to be able to
decrease anode surface area beyond a factor or 2 or 3.  This still is going
to require a significant amount of cathodic protection anode surface.
And now we need a power supply.

	Do you think the above is practicable for an automobile?

	Okay, now we have one more dragon to slay: anodic protection to
passivate the surface of the steel using an impressed current.  This is
much more difficult to achieve since not only will we will still require
auxiliary electrodes (in this scheme called cathodes), but we will require
VERY precise potential control using one or more reference electrodes
with electronic potentiostatic control.  Anodic protection is tricky,
because if the potential goes amok, it cause greatly accelerate corrosion
rates over what would occur without protection in the first place.

	Do you really think the above is practicable for an automobile?

	What you fail to realize is that while cathodic and anodic protection
schemes are indeed effective, they have limited usefulness.  These methods
are excellent for buried pipelines, buried tanks, above ground tanks, and
shipboard use - but they are generally difficult if not impracticable to
implement for other applications.  The successful applications all have one
thing in common which is NOT found in an automobile: the object to be
protected is ALWAYS in contact with an electrolyte which covers most if
not all of the surface to be protected!

> OH, you want a specific example.  Ok, first I buzz down to your house with
> my nibblers, cut a hunk of sheet metal out of your fender, bring it back
> here and put it upright in a largish beaker.  Next, I pour about 75 g of
> 10 mesh calcium chloride/sodium chloride (aka road salt) into the
> bottom.  Finally, I carefully pour some water (tap or distilled) into
> the beaker.  Viola, one steel-steel concentration cell using materials
> common on our nation's highways.  Happy?

	The above is not - I repeat NOT - a "steel-steel concentration cell".
While some concentration gradients may exist in your example, they pale
by comparison to what is really going on: a simple chemical reaction as
described in some detail in the beginning of this article.

> Well, without going into the gorey details, the fact is that plain
> diffusion is very slllooowww.  If I take a perfectly still beaker of
> liquid (at a uniform temperature to avoid bouyancy convection) and add a
> second liquid extremely carefully, and keep it where the temperature
> will remain uniform and as little vibrations as possible will shake it, it
> will take weeks, if not months, for the solution to come to chemical
> equilibrium.  Yes, this experiment has been performed several times. 
> Mixing is such an important aspect of chemical engineering that books
> are devoted to it.  It is not a trivial operation. 

	So what does the above have to do with the price of tea in China?

> You really should proof read, Larry.  Galvanizing is cathodic
> protection.  So, that's example one.
> Further,
> galvanizing works only because it is cathodic protection.

	I told you once, I told you twice, and I'm telling you again
that the primary effect of galvanizing is NOT cathodic protection, but
it is physical isolation through cladding.

> What I proposed is all very
> reasonable and practical.  I wouldn't say it's the best idea (I would
> just indulge in some Rust-Oleum),

	Finally something I agree with!

> >Then tell me about what will happen after the magnesium gets wet the
> >first time, and a solid layer of magnesium oxide forms on its surface. 
> 
> Larry, this is your best example of leaping without looking.  Magnesium
> is frequently used for sacraficial anodes on ships and inside water heaters.
> Do you think elves come along and scrape off the oxide?  Wrong, the oxide
> layer is not sufficient to prevent further corrosion of the magnesium
> (this is why we don't use aluminum as a sacrifical anode, it does form
> an excellent oxide layer).

	There is something which you are overlooking.  Magnesium works
just fine - when it has: (1) a significant surface area as compared to
the area to be protected; and (2) even more important - when it is
continuously immersed in the electrolyte!  Neither of these two events
are going to occur when bolting a magnesium block to a tire well in a
trunk, but they DO occur in your example of water heaters and ships.
Furthermore, another reason why I mentioned the oxide coating is that it
makes it difficult to achieve a low resistance electrical connection to
the metal to be protected.

> So, Larry, sorry I had to bust your bubble; but, I just can't let your
> brand of dis-information go by.

	Tell me about your grandiose plans to eliminate corrosion after
you've obtained your degree and worked in the real world of industry for
a few years.

<>  Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp.
<>  UUCP   {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry
<>  TEL  716/688-1231 | 716/773-1700  {hplabs|utzoo|uunet}!/      \uniquex!larry
<>  FAX  716/741-9635 | 716/773-2488     "Have you hugged your cat today?"