🚀 go-pugleaf

I'm glad to see we've got this down to a dull roar.

In article <3235@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes:

>1.	General Chemical Attack - This is THE primary mechanism behind
>	corrosion in automobiles, and is THE most common cause of any
>	corrosion of man-made objects.  As I pointed out in my previous
>	article, strictly speaking this is an electrochemical reaction,
>	but to think of it as such can often be seriously misleading to
>	many people since there are not two readily identifiable electrodes
>	or a readily identifiable source of electrical energy.  By the
>	expresion "readily identifiable" I refer to being understood
>	by people who do not have a good working knowledge of chemistry.

Larry, this is the whole heart of corrosion engineering.  Since it is
virtually impossible to prevent steel in coming in contact with water
and air, corrosion engineering is to prevent the formation of the
suitable conditions for this reaction to go.  ALL of the various
mechanism you listed below are stil general chemical attack.  The
only difference is that a particular reason has been fingered for
supplying the activation energy (galvanic couples, pits, stress, etc.).
It may not be easy to identify the electrodes, but one must if there is
any hope in controlling corrosion.

Thus, corrosion is corrosion.  Anything is a variation on the same theme.

>5.	Stress Corrosion - Somewhat self-explanatory name, and not at all
>	significant as a cause of automobile body rust.

Yes, this is a significant corrosion mechanism is *some* cars.  Did you
every see a body panel corrode along the character lines and folds
(good example:  GM Suburban)?

>6.	Pitting Corrosion - A highly localized concentration cell phenomenon
>	which causes - what else? - pits in the metal.

This is also significant mechanism in body panels where either defects or
corrosion is left during manufacture.  Since I've worked in the body
industry, I've seen this.

>	I agree with this!  [I say, we're making REAL progess now!]  I have
>also "considered" cathodic protection for this problem, and rejected it as
>being impracticable.  The point of my articles is to offer this bottom-line
>opinion to those readers who do not have the kwowledge and experience to
>evaluate the applicable issues themselves. 

Well, we not making progress, just getting things clearer.

The bottom line is that other methods are better, not just practical.  There
are situations where cathodic protection is practical (and I supplied one).
The reason it's not the best choice is that (1) paint is cheap and easy
to apply, (2) since we can tolerate some corrosion, we can accept paint
which supplies most imperfect protection, (3) the area we are considering
is easy to get to so application and maintainence is no problem.

However, (3) points to situations where paint would be dangerous.  Since
a flaw in paint worsens corrosion, we shouldn't use paint where there
isn't a back-up protection mechanism or where we can't easily find and
correct the flaw.  At this point cathodic protection comes into play
(and is used extensively as galvanized coatings).  I still see no reason
to completely reject the use of sacrafical anodes.

>	But it ain't that easy to be _effective_.  First, you have to
>fasten the zinc to the body, which is going to require drilling holes
>in the body.  Second, you are going to have to use a suitable fastener,
>which is an invitation to further problems.  Third, if the resultant
>connection does not have a damn low resistance, not only will the scheme
>NOT work, but it can readily cause further corrosion problems resulting
>from the fastening method.

Well, drilling holes is not a problem unless it's an outer class A body
panel (your car already has thousands of holes in it).  Next, a steel
nut and bolt should work very nicely.  The keep the connection clean, I
would use a conductive grease between the parts.

These problems are not insurmountable.  I agree that in some situations,
these concerns would probably make cathodic protection the sub-optimal
choice, but they are by no means barriers.

>	I hate to say this - since I don't want to provide fuel to use against
>my arguments :-) - but the pH of road salt electrolyte solutions tends to
>be alkaline, around a pH of 8 like that of seawater.

Every hear of acid rain? :-)

>	But unless the metal traverses both solution layers, there will be
>no potential difference across the metal.  While this mechanism of corrosion
>is common in boilers and chemical process equipment, the required conditions
>simply don't exist in the automobile underbody (except, perhaps, for some
>crevices, but this is only a very small part of the problem).

Well, crevise corrson is not a small part of the problem.  Crevise corrosion
occurs in the joints in the automotive unibody.  If these joints rust apart,
the car is junk.  Further, corrosion begets more corrosion; thus, once a 
good start is made in the crevice, this will spread the corrosion into the
homogeneous areas.

The point I've been making is that it is *very* easy to set-up concentration
gradients.  Once the corrosion is started, it will continue very easily in
conditions that would be unable to start corrosion on vigin steel.  Since
corrosion engineering is preventing your reaction from getting started,
this is a valid concern.

>Engineering students are MUCH more dangerous. :-) :-) :-)

Well fortunately: (1) I *am* an engineer, (2) you supplied smileys :).

Enough of this,

Jim Kweeder
kweeder@sun.soe.clarkson.edu