[Diy_efi] Possible method of partial repair of damaged O2 sensor (was designed for O2)

[Jay Wallace]

At 01:40 AM 03-06-05, you wrote:
>At 05:07 AM 3/06/05, you wrote:
> >I'm guessing that at least some of the Pb from leaded fuel probably 
> becomes an oxide (PbO) in the exhaust stream that reacts with the ZrO2 of 
> the sensor to form PbZrO3, a really stable compound that is not very 
> volatile or affected by oxygen at high temperatures. If the exhaust is 
> really reducing then it could be Pb or Pb2O that your oxygen rich torch 
> cycle would convert to PbO and...
>
>Wouldnt there be a probabilistic assesment of the nature of this reaction 
>based on
>the period of exposure in conjunction with the reaction energy, I mean 
>would there
>be enough heat for the period of time to make PbZr03 adhere strongly and 
>make it
>robust enough for permanent damage, what is the physical property of that 
>compound ?

WARNING: boring non-efi technical stuff follows -  you have been warned <G>

Mike,
You raise some interesting questions (at least for me, a ceramist who has 
worked with ZrO2 and PbZrO3) The fact that the sensor is affected by the 
presence of Pb would seem to indicate that there is some kind of reaction. 
The details of any reaction, however, are in question. I would guess that 
the kinetics of the reaction are rapid enough at 600 C that a few hours 
would be adequate for reaction to occur. When the volume of the reaction 
product is similar to that of the reactants, you can get a protective 
coating similar to the surface oxidation of aluminum. Here the molar volume 
of PbZrO3 is much larger than that of ZrO2 (ignoring the PbO vapor) so it 
may not be a stable coating. Also, any difference in coefficient of thermal 
expansion between ZrO2 and PbZrO3 would tend to spal off a coating when 
thermally cycled (don't have any data here for PbZrO3) Thermal cycling 
would happen during every startup - shutdown so I don't see the torch as 
offering much new.

The point that I was trying to make was that PbZrO3 is stable in a strongly 
oxidizing environment at high temperatures - it doesn't volatilize or 'burn 
off' as it is already a stable oxide.



>I did recall some small whitish flakes fly off when I turned up the heat, they
>gave off a light purplish or violet hue in the flame but there werent many.
>
> >Answer to your question: No.
> >
> >Or, a more socially acceptable answer: IMHO I doubt it.
>
>Is the stable compound fragile, I wonder if the pre-treatment with a propane
>torch and combination with vibration environment in normal use later
>might be enough to provide some minimalist rejuvenation ?
>
>Is the PbZr03 acting as a short or just occluding the sensor area, so a higher
>impedance output results. ie. If a lower impedance output is expected
>and the pre-amp designed accordingly and contamination occurs the output
>will fall away. If a much higher impedance pre-amp were used instead would
>the contamination be less likely to affect the output in that some area 
>occlusion
>could be tolerable ?


If there is PbZrO3 on the surface it would act as an oxygen barrier - it is 
not an ionic oxygen conductor like ZrO2. Since the ZrO2 sensor is acting 
like a battery (Nernst cell) conducting oxygen ions from one side to the 
other driven by the difference in oxygen content (partial pressure) between 
the inside and outside, any ZrO2 surface that is free of PbZrO3 should be 
adequate to produce the cell voltage. Current would scale with surface area 
but, since  high impedance amps are the rule rather than the exception, 
this probably would not be a problem until the bitter end.

All of this discussion assumes that there is a surface reaction rather than 
diffusion down grain boundaries. If the grain boundaries are coated with 
PbZrO3 stopping oxygen conduction, the time to sensor death would probably 
be much shorter.

Are you still awake? (and I don't mean the time zone difference)

I'll try to do a literature search and see if I find the real answer.

Jay