Cracking in FRP Road Tankers

[George77577]

We have an ongoing issue with cracking around nozzle attachments in over the road filament wound tankers. Only see this in sodium hypochlorite service. Appears to be predominate in the high stress areas such as nozzles and to some extent head attachment.

1. Road tankers have very high dynamic loads but since we don't see this in HCL have to assume its related to Hypo in some way.

2. We tried a higher elongation resin...onset of problems appears sooner. So assume resin is less resistant and effect is corrosion related but resin matrix shows no obvious sign of degradation.

3. Surface veils may play a role and increase rigidity = more susceptiple to cracking in high stress areas?

4. Hypo is unstable and possible corrosion effect is related to the E-glass fibers being attacked once veil layers are breached.

5. Degree of cracking and types vary from small micro to larger but only in Hypo service. Anyone have any ideas on the root cause? We have onboard samples and they provide some direction but not conclusive. Its been a battle for some years still trying to uncover.

Pictures show typical effect. Possible stress induced during cure due to shrinkage? Laminate is .180" could be reduced to .125" ? Not a stuctural laminate just corrosion barrier.

Any comments?

 

[George77577]

Another picture

 

[Demon3]

The sodium hypochlorite is attacking (oxidizing) your plastic. As you suspect, once the skin is breached, the glass fibres provide a pathway to wick the sodium hypochlorite into the plastic. This removes the bonding between the polymer and glass so strength is lowered dramatically and polymer degradation is accelerated because the glass fibers help the sodium hypochlorite penetrate into the bulk polymer.

This is a chemical effect and can only be tackled by using different chemistry. Either surface coating your existing plastic or changing to a new one.

Chris DeArmitt PhD FRSC CChem

http://www.phantomplastics.com

Consultant to the plastics industry

 

[George77577]

Demon,

I agree with your theory that the skin is breached but the solution seems to be in preventing the breach. What are your thoughts on exotherm induced stress into the laminate? The typical exotherm is around 120 - 140F @ .200".

If we can determine the cause of the cracks then can possibly changing to a non-glass fiber. Carbon failed in a Dow study but has outperformed E-glass and Basalt in our own study.

 

[Demon3]

To test the thermal shock theory you could heat and cool it fast a couple of times and see if similar cracks appear. It is only heated like that when there is sodium hypochlorite in there? You said it only fails with sodium hypochlorite so if there are other instances when it's heated but doesn't fail then we're back to it being the sodium hypochlorite and not the heating.

Chris DeArmitt PhD FRSC CChem

http://www.phantomplastics.com

Consultant to the plastics industry

 

[PanelGuy]

The bigger issue I think George is alluding to is the thermal induced microcracking a the fiber/resin interphase. A high exotherm may cause a problem with this. Also, this could contribute to the failure by creating mircro breache in a veil...

hmmm...Run a soak test on the laminate, the resin, neat, and the fibers alone. See where the issue really is. A coule of hours or days...just observe and it will tell you wehre to go. My guess is that it is more of an issue with the flexibilizers in the resin. If you went to a higher elongation resin perhaps the flexibilizer was more reactive (in the case of vinyl esters this is probably true)

I doubt the glass itself is a problem though...
In the end I see two main paths
1 - find a flexible sealant and seal the inside of the tank...extra around the bosses. I am thinking a brush on rubber (silicone?) coating but the chemistry needs to be clear first.
2 - change your resin, I don't mean vary, I mean go to a different chemical family. This could be tested with the soak tests ads outlined above.

So yeah, I agree with Demon...

 

[corrosionman]

George77577

A road tanker is a pressure vessel - -It will undergo a regular pressure test and also, some tankers are emptied by pressurising the inside and forcing the contents up a vertical pipe. Can these cracks be caused by internal tank pressure, either deliberate or accident ?? Are these branches / nozzles correctly designed - - they do look like cracks from excess pressure .
cheers
CM

 

[George77577]

CM you are correct. Normal operating pressure is 30-35 psi. However, the process is identical in all units built with slight variations of course due to the nature of the material. Over a broad range of units built we ONLY see this in sodium hypochlorite service. My guess is SCC.... we are inducing stress into the laminate in the form of exotherm and too rapid post cure. The corrosion is possibly a function of the unstable environment created by decomposed hypo. Products of decomposition are sodium chloride or chlorinated brine and possibly hypochlorous acid. Both are known to be extremely corrosive but due to the nature of the beast we have no way to test for this.

Operators that routinely rinse after every load don't seem to have the problem. That would lead creedence to the decomposition theory. Novolac vinyl esters do much better in oxidizing environments so possibly explains sample showing good novolac performance.

I still have my doubts we have this issue nailed but think we are getting closer. Controlling exotherm is a start but doubtful that it will be the only answer.

See the attached pics of the samples on board installed during original mfg and see what you think.